[ { "name": "1Password AI Agent Identity Kit", "category": "Agent identity & access", "description": "A developer kit layered on 1Password vaults and SDKs that gives a software agent a verifiable identity tied to a workload, issues short-lived agent tokens instead of shared API keys, scopes what the agent may access and links agent actions to audit trails. It sits within the broader 1Password Unified Access platform.", "url": "https://1password.com/solutions/ai-agent-identity", "question": "Show an agent receiving a short-lived credential tied to its workload identity, using an approved secret without exposing it, and the audit trail linking the action to a user.", "reviewed": "2026-09-06" }, { "name": "Aembit Workload IAM", "category": "Agent identity & access", "description": "A workload identity and access platform that attests the client environment of a workload or AI agent, evaluates a policy at request time, then injects short-lived credentials into the outbound API call so the workload never stores a secret. Discovery and inventory of existing non-human identities is not evidenced.", "url": "https://aembit.io/product/", "question": "Demonstrate an AI agent obtaining a 15-minute credential through client attestation and policy evaluation, and show the access log for a denied request outside business hours.", "reviewed": "2026-09-06" }, { "name": "agentgateway", "category": "Runtime authorization & controls", "description": "Open-source proxy that carries service, LLM-provider, MCP and agent-to-agent traffic in one data plane, letting platform teams route agent tool calls through a single enforcement and audit point. Project documentation covers MCP proxying and routing configuration; tool-level RBAC and token exchange are described mainly on vendor product pages.", "url": "https://agentgateway.dev/", "question": "Show me an MCP server proxied through agentgateway with tool-level RBAC denying one specific tool call, and the audit record it produces.", "reviewed": "2026-09-06" }, { "name": "AgentOps", "category": "Observability & traceability", "description": "MIT-licensed monitoring platform for AI agents that records each program execution as a session and visualises a waterfall of LLM calls, action events, tool calls and errors, with prompt and completion inspection, event-type breakdowns, timings and LLM spend tracking. Self-hostable; no documented access control or retention.", "url": "https://docs.agentops.ai/v2/introduction", "question": "Show me the session waterfall for one agent execution, including tool calls, errors and the exact prompt and completion for a chosen LLM call.", "reviewed": "2026-09-06" }, { "name": "AGNTCY Identity", "category": "Agent identity & access", "description": "Apache-2.0 project from the Linux Foundation-hosted AGNTCY effort that issues and verifies identities for AI agents, MCP servers and multi-agent systems using verifiable credentials and badges, and can onboard existing identities from providers, A2A agent cards or W3C DIDs. Deployment and adoption maturity are unproven.", "url": "https://github.com/agntcy/identity", "question": "Demonstrate issuing a verifiable badge to an MCP server and having a relying agent verify it, including how revocation and key storage work.", "reviewed": "2026-09-06" }, { "name": "AI Incident Database (AIID)", "category": "AI risk & compliance management", "description": "Open, community-contributed catalogue of real-world AI harms and near-harms maintained by the Responsible AI Collaborative, searchable on the web and queryable through a public read-only GraphQL endpoint, with taxonomies for classification. Records are curated press-based reports, so coverage is uneven and not an authoritative incident register.", "url": "https://incidentdatabase.ai/", "question": "Show me querying the GraphQL endpoint for incidents involving autonomous agents or LLM assistants and how you would use them in our AI risk taxonomy.", "reviewed": "2026-09-06" }, { "name": "Akeyless SecretlessAI", "category": "Agent identity & access", "description": "Akeyless brokers AI agent connections through its Gateway so agents reach databases, SaaS, cloud services and legacy systems without receiving passwords, API keys or certificates; access is created on demand, scoped by policy, revoked afterwards and recorded in a central audit trail. Requires routing agent traffic through the Gateway.", "url": "https://www.akeyless.io/secrets-management/secretless-ai-agents/", "question": "Demonstrate an MCP-based agent querying a production database through the Akeyless Gateway with no credential ever entering the agent's context, plus the audit record.", "reviewed": "2026-09-06" }, { "name": "Amazon Bedrock AgentCore", "category": "Agent building & orchestration", "description": "AWS's set of composable services for running agents built with any framework, covering serverless execution with isolated sessions, persistent memory, a gateway that turns APIs into MCP tools, identity and credential management, and built-in observability. Documented governance is strongest on identity and authorization rather than approvals.", "url": "https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/what-is-bedrock-agentcore.html", "question": "Show me an AgentCore Gateway exposing two Lambda tools where the agent's inbound and outbound authorization is verified per request through AgentCore Identity.", "reviewed": "2026-09-06" }, { "name": "Amazon Bedrock Guardrails", "category": "Agent security & threat detection", "description": "Configurable safeguard service inside Amazon Bedrock that evaluates user inputs and model responses against content filters, denied topics, sensitive information filters and word filters, including a prompt attack category. It can be applied at inference or via a standalone API, but does not authorize agent tool calls.", "url": "https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails.html", "question": "Show me a Bedrock guardrail blocking a prompt attack and masking PII for a Bedrock Agent, then show the same guardrail invoked through ApplyGuardrail for a non-Bedrock model.", "reviewed": "2026-09-06" }, { "name": "Amazon SageMaker Model Registry", "category": "Model lifecycle & governance", "description": "AWS service for cataloguing production models as versioned model packages in model groups, with metadata, lineage, a staging construct, approval status and CI/CD deployment. Integrated SageMaker Model Cards add intended use, risk rating and evaluation records, versioned immutably on edit.", "url": "https://docs.aws.amazon.com/sagemaker/latest/dg/model-registry.html", "question": "Demonstrate a model package moving through staging with approval status changes, and show the linked model card version history for the same model.", "reviewed": "2026-09-06" }, { "name": "AppOmni Agent Inventory", "category": "Agent discovery & inventory", "description": "Capability of AppOmni's SaaS security platform that surfaces AI agents running inside connected SaaS tenants - such as Salesforce Agentforce, ServiceNow Now Assist and Microsoft 365 Copilot - including agents enabled without security review, with their declared tools, identities, permissions and over-privilege findings. Discovery depends on AppOmni's SaaS API connections.", "url": "https://appomni.com/ai-security/agent-inventory/", "question": "Demonstrate listing every Agentforce and Now Assist agent in our tenants, flagging which ones were enabled without approval and which hold write or destructive permissions.", "reviewed": "2026-09-06" }, { "name": "Arcade", "category": "Runtime authorization & controls", "description": "Actions runtime that brokers agent tool calls to SaaS systems, handling OAuth and user token storage and applying per-action authorization so an agent acts within both the user's and its own scope. Pre- and post-tool-call hooks allow blocking or redaction, though the hosted tool catalogue is vendor-maintained.", "url": "https://www.arcade.dev/", "question": "Show me an agent calling a Gmail tool through Arcade where the user consents via OAuth, a pre-tool-call hook blocks a send action, and the audit log records the decision.", "reviewed": "2026-09-06" }, { "name": "Arize AX", "category": "Observability & traceability", "description": "Commercial AI engineering platform that captures traces of agent runs including inputs, outputs, tools and costs, groups them into sessions, visualises agent trajectories, and runs online and offline evaluations with human annotation. Arize also maintains Phoenix, a local-first open-source tracing and evaluation package.", "url": "https://arize.com/docs/ax", "question": "Show me an agent trajectory visualisation for a failed multi-step run and the trace-level evaluation scores attached to it.", "reviewed": "2026-09-06" }, { "name": "Asenion", "category": "AI risk & compliance management", "description": "AI governance, risk and security platform formed when Canada's Fairly AI acquired Sweden's anch.AI in June 2025 and rebranded as Asenion. It markets automated controls that continuously assess and test AI systems and agents against frameworks including the EU AI Act, ISO/IEC 42001 and NIST AI RMF.", "url": "https://www.asenion.ai/", "question": "Demonstrate one automated control test running against a production agent, and show the tamper-resistant assurance record it produces.", "reviewed": "2026-09-06" }, { "name": "Astrix Agent Control Plane", "category": "Agent identity & access", "description": "Astrix discovers AI agents, MCP servers, service accounts, OAuth apps, API keys and other non-human identities across cloud, SaaS, CI/CD and vaults, maps each to a human owner in an identity graph, and applies agent policies plus onboarding and offboarding actions. Enforcement depth outside integrated platforms is unclear.", "url": "https://astrix.security/product/secure-ai-agents/", "question": "Show me the identity graph for one shadow agent — its NHIs, credentials, reachable resources and owner — then apply a policy that blocks it and offboard it.", "reviewed": "2026-09-06" }, { "name": "Auth0 Auth for GenAI (Token Vault & delegated authorization)", "category": "Agent identity & access", "description": "Developer-facing authorization service from Auth0 that lets an AI agent call first-party and third-party APIs on behalf of an authenticated user using OAuth scopes, with Token Vault obtaining, storing and refreshing external API tokens. It governs user-delegated agent access, not enterprise-wide agent inventory.", "url": "https://auth0.com/ai/docs/intro/delegated-authorization", "question": "Demonstrate an agent obtaining a Google and Slack token from Token Vault after user consent, and show what happens when the agent requests a scope the user never granted.", "reviewed": "2026-09-06" }, { "name": "Azure AI Content Safety Prompt Shields", "category": "Agent security & threat detection", "description": "Microsoft API in Azure AI Content Safety that analyses user prompts and supplied documents for adversarial instructions, returning attack-detected flags for direct user prompt attacks and indirect attacks embedded in external content. It is a detection endpoint that applications must act on rather than an enforcement gateway.", "url": "https://learn.microsoft.com/en-us/azure/ai-services/content-safety/concepts/jailbreak-detection", "question": "Show me a shieldPrompt call flagging an indirect injection hidden in an uploaded document, and how the calling agent enforces a block based on that response.", "reviewed": "2026-09-06" }, { "name": "BABL AI audits and AI & Algorithm Auditor Certification", "category": "Advisory & assurance services", "description": "Boutique firm offering independent third-party AI audits and responsible AI consulting, plus a five-course AI and Algorithm Auditor Certification for practitioners ending in a capstone and exit exam. The firm states its audits follow assurance-engagement style practice but names no accrediting body for either the audits or the credential.", "url": "https://babl.ai/", "question": "Demonstrate how a BABL independent third-party audit engagement is scoped and what assurance wording appears in the final report you issue to a client's stakeholders.", "reviewed": "2026-09-06" }, { "name": "BigID AI Security & Governance", "category": "Data governance & privacy", "description": "BigID's AI-oriented module inside its data security platform. It inventories AI models, agents, copilots, prompts, vector stores and pipelines, classifies the data feeding training, retrieval and inference, and maps lineage of that data. Claims come from vendor product pages; BigID's technical documentation is not publicly reachable.", "url": "https://bigid.com/ai-security-governance/", "question": "Demonstrate how BigID builds an AI asset inventory that links a vector database to the sensitive datasets embedded in it and the identities allowed to query it.", "reviewed": "2026-09-06" }, { "name": "Boomi Agent Control Tower", "category": "Agent discovery & inventory", "description": "Registry and monitoring console within Boomi Agentstudio where organisations register, tag, monitor and manage AI agents across providers - agents built in Boomi and third-party environments such as Amazon Bedrock - via configured provider connections. Agents appear because a provider account is connected, so unconnected environments remain invisible.", "url": "https://help.boomi.com/docs/Atomsphere/Platform/Agent_Control_Tower", "question": "Show me registering an Amazon Bedrock provider account and then disabling one of its agents directly from Agent Control Tower, including the trust level and tags applied.", "reviewed": "2026-09-06" }, { "name": "Braintrust", "category": "Evaluation & testing", "description": "Developer platform for scoring AI application output: teams log traces from their app, curate datasets from human feedback, run experiments against scorers, then watch production logs. Documentation covers project-level access control and a customer-controlled data plane, but audit logging and retention behaviour are not described.", "url": "https://www.braintrust.dev/docs", "question": "Show me an experiment run comparing two prompt versions against a dataset built from human feedback, and who in my organization can view the underlying logs.", "reviewed": "2026-09-06" }, { "name": "Britive ARC (Agentic Runtime Control)", "category": "Agent identity & access", "description": "Britive registers each AI agent as its own identity with a human owner and permitted access profiles, authenticates it via SPIFFE SVID, OIDC federation or API tokens on every request, and grants just-in-time privileges that are created and destroyed per task so no standing credential remains. Cloud-target coverage varies.", "url": "https://www.britive.com/platform/agentic-ai-identity-security", "question": "Register an agent with an owner and job function, then show a just-in-time credential created for one task and destroyed at task end, with the access record.", "reviewed": "2026-09-06" }, { "name": "BSI ISO/IEC 42001 certification", "category": "Advisory & assurance services", "description": "Third-party certification of an organisation's AI management system against ISO/IEC 42001, offered by BSI alongside pre-certification gap assessment and training. BSI states it holds UKAS, RvA and ANAB accreditation for this scheme. Certification covers management-system conformity, not the performance or safety of individual AI models.", "url": "https://www.bsigroup.com/en-US/products-and-services/standards/iso-42001-ai-management-system/", "question": "Show me the audit stages, sampling approach and evidence you require to certify an AI management system covering generative AI agents in production.", "reviewed": "2026-09-06" }, { "name": "Camunda Agentic Orchestration", "category": "Agent building & orchestration", "description": "German process-orchestration vendor's capability for embedding AI agents into BPMN processes, exposing BPMN activities inside ad-hoc sub-processes as LLM-selectable tools while human tasks and deterministic rules stay in the same model. Fine-grained authorization exists at cluster level; audit logging for agent decisions is not documented.", "url": "https://docs.camunda.io/docs/components/agentic-orchestration/agentic-orchestration-overview/", "question": "Demonstrate a BPMN process where an AI agent selects tools inside an ad-hoc sub-process but a human task must complete before the payment activity runs.", "reviewed": "2026-09-06" }, { "name": "Check Point AI Agent Security (formerly Lakera Guard)", "category": "Agent security & threat detection", "description": "Runtime protection for AI agents that inspects prompts, reference material, tool responses and tool descriptions for injections and manipulation, applies tool allow and deny lists, and flags actions outside an agent's mandate. Also builds an inventory of agents and connected MCP servers across supported agent platforms.", "url": "https://docs.lakera.ai/guard", "question": "Show me Check Point AI Guardrails detecting an injection hidden in a tool response and blocking the resulting tool call, plus the agent inventory entry for that agent's MCP servers.", "reviewed": "2026-09-06" }, { "name": "Checkmarx AI Inventory and AI-BOM", "category": "Agent discovery & inventory", "description": "Capability inside Checkmarx One that inventories AI components by scanning source code and configuration files in connected repositories, cataloguing models, agents, MCP servers, AI libraries and SDKs on every commit, then emitting an AI-BOM and enforcing policy in pull requests and CI/CD. Scope is the software pipeline, not employee tool usage.", "url": "https://checkmarx.com/ai-bom/", "question": "Show me an AI-BOM generated from one of our repositories listing every detected model, agent and MCP server with the file and commit where it was found.", "reviewed": "2026-09-06" }, { "name": "Cisco AI Defense AI BOM", "category": "Agent discovery & inventory", "description": "Open-source scanner from Cisco's AI Defense team that inspects codebases, container images and cloud environments to produce an AI bill of materials listing models, agents, tools, MCP servers and clients, datasets, prompts, guardrails and secrets. Detection is static analysis plus catalog matching, with optional LLM-based enrichment.", "url": "https://github.com/cisco-ai-defense/aibom", "question": "Demonstrate running the scanner against one of our Python repositories and a container image, and show the resulting AI-BOM entries for agents, tools and MCP servers.", "reviewed": "2026-09-06" }, { "name": "Cisco MCP Scanner", "category": "Agent security & threat detection", "description": "Apache-2.0 Python tool from Cisco's AI Defense group that scans MCP servers, their tools, prompts and resources plus server source code, combining YARA rules, LLM-based analysis and Cisco's hosted inspection API, and audits dependencies and bundled binaries. Full detection depth depends on optional third-party services.", "url": "https://github.com/cisco-ai-defense/mcp-scanner", "question": "Show me a scan of an untrusted MCP server package that flags a docstring-versus-implementation mismatch, and which findings required the Cisco AI Defense API.", "reviewed": "2026-09-06" }, { "name": "ClearML Model Registry", "category": "Model lifecycle & governance", "description": "Model registry inside the open-source ClearML MLOps toolchain: models are logged automatically or manually from training tasks, catalogued per project with metadata and metrics columns, queryable by name, tag or metadata, with lineage and CI/CD triggers on registry events such as tagging or publishing.", "url": "https://clear.ml/docs/latest/docs/model_registry/", "question": "Demonstrate an automatic model registration from a training task, then trigger a CI/CD pipeline when that model is tagged for production.", "reviewed": "2026-09-06" }, { "name": "Collibra AI Command Center", "category": "AI risk & compliance management", "description": "Collibra's AI governance product, evolved from Collibra AI Governance, registering AI use cases, models, model versions and agents as governed assets with lifecycle stages, compliance assessment templates and a per-system trust score. Fits organisations already using Collibra for data governance.", "url": "https://www.collibra.com/products/ai-command-center", "question": "Show me an AI use case linked to its model versions, agents and datasets, plus how the trust score changes when documentation or lifecycle status degrades.", "reviewed": "2026-09-06" }, { "name": "Confident AI", "category": "Evaluation & testing", "description": "Hosted quality platform from the DeepEval maintainers. It captures each LLM call as a trace with inputs, outputs, tool calls, latency, token cost and metadata, converts flagged traces into evaluation datasets, and runs metric-based regression tests on pull requests. Governance controls are not documented.", "url": "https://documentation.confident-ai.com/", "question": "Show me how a failing production trace becomes a dataset case that then blocks a pull request when the metric regresses.", "reviewed": "2026-09-06" }, { "name": "ContextForge AI Gateway (mcp-context-forge)", "category": "Runtime authorization & controls", "description": "Apache-2.0 gateway, registry and proxy from IBM that federates MCP servers, A2A agents and REST or gRPC APIs behind one endpoint, adding authentication, rate limiting, input validation and OpenTelemetry tracing. It centralises tool discovery for MCP clients but does not itself provide threat detection or sandboxed execution.", "url": "https://github.com/IBM/mcp-context-forge", "question": "Show me ContextForge federating two MCP servers plus a REST API behind one endpoint, with a user-scoped OAuth token and rate limit applied to one tool.", "reviewed": "2026-09-06" }, { "name": "Coval", "category": "Evaluation & testing", "description": "Simulation and evaluation platform for voice and chat agents. It generates large volumes of simulated callers with accents, interruptions, noise and policy traps, scores production calls in real time on resolution and safety, detects regressions after prompt or model changes, and routes failures to human reviewers.", "url": "https://docs.coval.dev/", "question": "Demonstrate a pre-launch simulation of a thousand callers against my voice agent, then show the production scoring that flags a regression after a prompt change.", "reviewed": "2026-09-06" }, { "name": "Credo AI Platform", "category": "AI risk & compliance management", "description": "Governance workspace where an enterprise records AI use cases, models, agents and third-party AI vendors, runs questionnaire-driven reviews, and tracks control and risk libraries with task assignment. A Python/TypeScript SDK writes the same objects programmatically. Evidence reviewed covers workflow structure, not independent verification of governance outcomes.", "url": "https://www.credo.ai/", "question": "Show me how a new agent use case moves from intake questionnaire through control review to a governance status a regulator-facing auditor could read.", "reviewed": "2026-09-06" }, { "name": "CrewAI", "category": "Agent building & orchestration", "description": "Open-source Python framework for orchestrating role-based agent crews, paired with CrewAI AMP, a hosted platform that deploys crews, exposes them via REST, streams events and shows execution traces. Human review is a framework-level pattern; platform documentation reviewed showed no approval console or audit-log feature.", "url": "https://docs.crewai.com/en/enterprise/introduction", "question": "Demonstrate a crew deployed on CrewAI AMP where a task requires human review before the next agent runs, and show the execution trace for that run.", "reviewed": "2026-09-06" }, { "name": "CrowdStrike AIDR Collectors", "category": "Agent discovery & inventory", "description": "Telemetry-collection layer of CrowdStrike's AI Detection and Response product. Collectors capture AI activity from the Falcon sensor (browser extension plus network inspection of desktop AI apps and coding assistants), standalone browser extensions, application SDK calls, and supported API gateways. Browser collectors log detections in report-only mode for model responses.", "url": "https://aidr-docs.crowdstrike.com/docs/aidr/collectors/", "question": "Demonstrate registering a Falcon endpoint collector and show which desktop AI applications and coding assistants network inspection surfaced that the browser extension missed.", "reviewed": "2026-09-06" }, { "name": "CSA STAR for AI", "category": "Advisory & assurance services", "description": "Cloud Security Alliance assurance program extending its STAR registry to AI services, with a Level 1 self-assessment against the AI Controls Matrix questionnaire, an automated validation option, and a Level 2 tier referencing third-party certification. Registry-based transparency for AI providers rather than a regulatory conformity assessment.", "url": "https://cloudsecurityalliance.org/star/ai", "question": "Demonstrate a completed AI-CAIQ submission for an agent platform and what the Valid-AI-ted scoring adds over a plain Level 1 self-assessment.", "reviewed": "2026-09-06" }, { "name": "Cycode AI & ML Inventory and AIBOM", "category": "Agent discovery & inventory", "description": "Application-security capability that discovers AI assets across the software development lifecycle - AI infrastructure, models, coding assistants, packages and associated secrets - by scanning connected repositories and pipeline systems, then producing an AI bill of materials for governance. Buyer is AppSec; it does not observe business users' AI tool usage.", "url": "https://cycode.com/blog/tackle-shadow-ai-with-cycode-ai-ml-inventory/", "question": "Demonstrate discovering which AI coding assistants and third-party models our developers introduced last quarter, and show the AIBOM entry with the repository and secret associations.", "reviewed": "2026-09-06" }, { "name": "Cyera AI Guardian", "category": "Data governance & privacy", "description": "Cyera's AI-focused extension of its data security posture platform. Its AI-SPM capability inventories AI models, applications, agents and knowledge bases including shadow AI, and links them to sensitive-data classifications produced by the underlying DSPM engine. Runtime protection is described at product level without public technical documentation.", "url": "https://www.cyera.com/platform/ai-spm", "question": "Show me the live inventory of AI agents and knowledge bases Cyera discovered in my cloud accounts and which sensitive data classifications each one touches.", "reviewed": "2026-09-06" }, { "name": "Databricks Unity Gateway", "category": "Runtime authorization & controls", "description": "Runtime governance layer inside the Databricks platform that routes and controls requests to models, agents, MCP services and AI tools using Unity Catalog privileges, service policies, rate limits and spend caps. Controls are strongest for assets governed by Unity Catalog rather than arbitrary external agent traffic.", "url": "https://www.databricks.com/product/artificial-intelligence/unity-gateway", "question": "Show me Unity Gateway restricting an agent to an approved subset of tools on an MCP service, with the service policy that blocks a request and the resulting inference log.", "reviewed": "2026-09-06" }, { "name": "Datadog Agent Observability", "category": "Observability & traceability", "description": "Datadog's AI observability product, documented as span kinds for LLM, workflow, agent, tool, task, embedding and retrieval. LLM spans record model and provider, session ID, inputs and outputs, tool calls, results and definitions, token and cost metrics, metadata and errors, inside an enterprise platform with audit trail, SAML SSO and access control.", "url": "https://docs.datadoghq.com/llm_observability/", "question": "Show me an agent span tree with tool calls and token cost metrics, and how retention plus RBAC apply to the captured prompt and response data.", "reviewed": "2026-09-06" }, { "name": "Datawiza Agent Gateway", "category": "Runtime authorization & controls", "description": "Identity-aware proxy positioned in front of MCP servers and APIs so that users and agents authenticate through an existing IdP before any tool call proceeds. Product material describes filtering tools/list discovery and gating tools/call invocation by policy; the fetched documentation site provides only high-level confirmation of that scope.", "url": "https://www.datawiza.com/mcp-gateway", "question": "Show me Datawiza filtering the tools/list response for a low-privilege group and denying a tools/call write action after validating the Entra ID token claims.", "reviewed": "2026-09-06" }, { "name": "Daytona", "category": "Runtime authorization & controls", "description": "Sandbox infrastructure for executing AI-generated code, offering per-sandbox isolation with a dedicated kernel, filesystem and network stack plus API keys, limits, network limits and audit logs. Buyers should note the public repository states core development moved to a private codebase and the repo is unmaintained.", "url": "https://www.daytona.io/", "question": "Show me a Daytona sandbox with network limits and audit logging enforced while an agent executes generated code, and clarify which components remain open source today.", "reviewed": "2026-09-06" }, { "name": "DeepEval", "category": "Evaluation & testing", "description": "Open-source Python framework for testing LLM applications: it runs metric-based scoring on test cases and on captured agent trajectories, including individual LLM calls, tool use, retrieval and sub-agent handoffs. It executes locally in developer or CI environments and provides no access control or audit records itself.", "url": "https://github.com/confident-ai/deepeval", "question": "Show me a CI test run that scores my agent's full trajectory, including whether the right tools were called with the right arguments.", "reviewed": "2026-09-06" }, { "name": "Deeploy", "category": "Model lifecycle & governance", "description": "Dutch platform that puts governance in the deployment path: models are registered from Git, MLflow, Databricks, Hugging Face or Azure registries, standardised assessments set a use-case risk score that selects applicable controls, role-based approvals gate deployment, and a gateway handles monitoring, guardrails and logging.", "url": "https://docs.deeploy.ai/", "question": "Show me a deployment blocked until the required roles approve, with the risk score, selected control framework and gateway monitoring for the approved version.", "reviewed": "2026-09-06" }, { "name": "Deloitte Algorithm Assurance", "category": "Advisory & assurance services", "description": "Named specialist assurance offering from Deloitte member firms that reviews client algorithms and the controls around them: definition, identification, classification, assessment, code review, stress testing, and monitoring. Aimed at audit committees and risk functions. Pages describe activities and objectives without naming an assurance standard or reporting format.", "url": "https://www.deloitte.com/au/en/services/audit-assurance/services/algorithm-assurance.html", "question": "Demonstrate how you classify an inventory of production algorithms and what your third-party code review report on one high-risk model actually concludes.", "reviewed": "2026-09-06" }, { "name": "Deloitte Trustworthy AI", "category": "Advisory & assurance services", "description": "Deloitte US consulting practice organised around a Trustworthy AI framework, sold as named workstreams covering AI strategy, risk management and governance, regulatory support, model risk management, and AI audit and assurance. The public pages describe offerings and control activities but publish no methodology, deliverable list, or fee information.", "url": "https://www.deloitte.com/us/en/what-we-do/capabilities/applied-artificial-intelligence/services/ethics-of-ai-framework.html", "question": "Show me the audit-ready evidence package your AI Risk Management and Governance engagement produces for a single high-impact agentic use case, from use-case tiering through monitoring KPIs.", "reviewed": "2026-09-06" }, { "name": "Descope Agentic Identity Hub", "category": "Agent identity & access", "description": "An identity provider for AI agents and MCP servers that registers OAuth clients (including dynamic client registration), records agentic identities created by user consent or tenant grants, and issues short-lived scoped credentials with policy-based access to protected APIs and MCP resources. Aimed at application builders rather than workforce IT.", "url": "https://docs.descope.com/agentic-identity-hub", "question": "Show me an autonomous agent registering via DCR, receiving a short-lived scoped token, and being denied an MCP tool call by an access policy — with the audit record.", "reviewed": "2026-09-06" }, { "name": "Dify", "category": "Agent building & orchestration", "description": "Open-source platform for building agents, agentic workflows, chatbots and RAG pipelines that can draw on private data and ship as web apps or APIs, with LLMOps monitoring and third-party tracing integrations. Documentation reviewed showed no approval workflow, role-based access control or audit-log capability.", "url": "https://github.com/langgenius/dify", "question": "Show me a Dify agentic workflow published as an API, and explain exactly what record exists of each model and tool call it made.", "reviewed": "2026-09-06" }, { "name": "Dioptra", "category": "Evaluation & testing", "description": "US NIST-built test platform for assessing trustworthy characteristics of AI models, providing a REST API, web interface and Python client to design, run and track reproducible experiments including adversarial red-team scenarios. Intended to support the Measure function of the AI Risk Management Framework rather than to certify systems.", "url": "https://github.com/usnistgov/dioptra", "question": "Demonstrate reproducing an adversarial robustness experiment in Dioptra and exporting the tracked evidence for an internal AI risk review.", "reviewed": "2026-09-06" }, { "name": "DNV ISO/IEC 42001 certification and AI assurance", "category": "Advisory & assurance services", "description": "DNV offers third-party certification of AI management systems to ISO/IEC 42001 plus an AI vendor capability assessment delivered as an independent third-party audit of an organisation's ability to develop and operate trustworthy AI and data-driven solutions. Neither page names an accreditation body for the AI scheme.", "url": "https://www.dnv.co.uk/assurance/Management-Systems/42001-ai-management-campaign/42001-ai-management/", "question": "Show me the topic-by-topic findings structure of an AI vendor capability assessment and how it differs from an ISO/IEC 42001 certification audit.", "reviewed": "2026-09-06" }, { "name": "Docker MCP Gateway", "category": "Runtime authorization & controls", "description": "MIT-licensed gateway from Docker that acts as a proxy between MCP clients and MCP servers, running each catalogued local server in an isolated container with restricted privileges, network access and resources, injecting credentials centrally and applying per-profile tool allowlists. It provides no semantic inspection of prompts or tool arguments.", "url": "https://docs.docker.com/ai/mcp-gateway/", "question": "Show me the MCP Gateway running two catalogue servers in isolated containers with a profile tool allowlist that hides one tool, plus the call trace for a blocked request.", "reviewed": "2026-09-06" }, { "name": "Domino AI Governance", "category": "Model lifecycle & governance", "description": "Governance layer of the Domino enterprise data science platform: an MLflow-based model registry with project- and deployment-scoped views, custom model cards, version management, RBAC over registered models and stage transitions, plus documented review steps for validation, ethical review, audit trails and stakeholder sign-off.", "url": "https://domino.ai/platform/ai-governance", "question": "Show me role-based control over stage transitions in the Domino model registry and the self-documenting evidence produced for a model review.", "reviewed": "2026-09-06" }, { "name": "DutyGraph", "category": "Work discovery & human delegation", "description": "Advisor-led workspace that maps a team's people, duties, tasks and software into a work record, then proposes agent delegation boundaries with a named accountable owner and review checkpoint. Intended for advisors and business sponsors. Currently an advisor pilot recruiting 5-10 companies; governance views are shown as a fictional sample.", "url": "https://dutygraph.com/landing/", "question": "Show a participant-reviewed task card reaching the advisor queue, and demonstrate how an unresolved approval owner stays visible in the company graph.", "reviewed": "2026-09-06" }, { "name": "E2B", "category": "Runtime authorization & controls", "description": "Apache-2.0 sandbox infrastructure that gives agents isolated cloud environments to execute model-generated code, process files and run tools, controlled from Python or JavaScript SDKs. Sandboxes bound where agent code runs, but E2B is an execution boundary rather than a policy decision point for tool authorization.", "url": "https://docs.e2b.dev/", "question": "Show me an agent running untrusted generated code in an E2B sandbox with filesystem and network limits, then show the same workload on a self-hosted E2B cluster.", "reviewed": "2026-09-06" }, { "name": "Entro Security NHI & Agentic AI Platform", "category": "Agent identity & access", "description": "Entro inventories non-human identities, secrets and agentic AI deployments across cloud, code, CI/CD, on-prem and SaaS, links each agent to the NHIs, entitlements and secrets it uses and to a human owner, and monitors agent behaviour for anomalies through its NHIDR detection engine. Credential issuance is not part of the evidenced scope.", "url": "https://entro.security/platform-ai-agents/", "question": "Show me an agent's NHI lineage — creator, secrets used, entitlements, resources touched — and a live NHIDR alert for anomalous agent behaviour.", "reviewed": "2026-09-06" }, { "name": "Enzai", "category": "AI risk & compliance management", "description": "UK-founded AI governance platform providing a system of record for AI systems, models, datasets and governance decisions, customisable intake by system type and risk level, assessment tracking against EU AI Act, ISO 42001 and NIST AI RMF, and dashboards summarising assessment results.", "url": "https://www.enz.ai/", "question": "Demonstrate a configurable intake for a high-risk system and how completed assessments roll up into the governance dashboard and shareable reports.", "reviewed": "2026-09-06" }, { "name": "Evidently", "category": "Model lifecycle & governance", "description": "Apache-2.0 Python library plus self-hostable platform for evaluating and monitoring data and AI systems, including tabular data quality and drift tests with statistical methods, declarative test suites usable in CI/CD, and dashboards tracking metrics and test results over time for deployed models.", "url": "https://github.com/evidentlyai/evidently", "question": "Show me a drift test suite failing in CI for a production model's input data, and how the result appears on the monitoring dashboard over time.", "reviewed": "2026-09-06" }, { "name": "EY Responsible AI services", "category": "Advisory & assurance services", "description": "EY advisory offering combining a Responsible AI framework, a GenAI governance framework, and a Responsible AI Readiness Assessment that scores an organisation's readiness to manage AI risk and coming regulation across six categories, plus stakeholder training. Public pages give category names but not the underlying scoring model.", "url": "https://www.ey.com/en_uk/services/ai/responsible-ai", "question": "Demonstrate the Responsible AI Readiness Assessment on one business unit and show the six-category scoring and the gap remediation plan it produces.", "reviewed": "2026-09-06" }, { "name": "Fiddler", "category": "Observability & traceability", "description": "Monitoring platform spanning traditional ML models, LLM applications and multi-agent systems. For LLM applications customers publish prompts, prompt context, responses and retrieved source documents; Fiddler generates trust and safety metrics, embeddings with UMAP visualisation and drift detection to support alerting and root-cause analysis.", "url": "https://docs.fiddler.ai/", "question": "Show me drift and trust-and-safety metrics for a RAG application, including the retrieved source documents behind a flagged response.", "reviewed": "2026-09-06" }, { "name": "garak", "category": "Evaluation & testing", "description": "Apache-licensed LLM vulnerability scanner maintained by NVIDIA. It fires static, dynamic and adaptive probes at a model or dialog system to test for jailbreaks, prompt injection, toxicity, data leakage and misinformation, logs each generation and detector verdict, and outputs a report with failure rates and hit logs.", "url": "https://github.com/NVIDIA/garak", "question": "Show me a garak scan of my chatbot with the probe-by-probe failure rates and the hit log for successful jailbreaks.", "reviewed": "2026-09-06" }, { "name": "Gemini Enterprise Agent Platform", "category": "Agent building & orchestration", "description": "Google Cloud's renamed Vertex AI agent stack, combining a managed agent runtime with sessions, a memory bank, example store and evaluation service, and support for ADK, LangGraph, LlamaIndex and A2A agents. Governance evidence centres on registration through the Gemini Enterprise app and built-in tracing rather than approval workflows.", "url": "https://docs.cloud.google.com/agent-builder/agent-engine/overview", "question": "Demonstrate deploying an ADK agent to the managed runtime, registering it in Gemini Enterprise, and inspecting its traces in Cloud Trace.", "reviewed": "2026-09-06" }, { "name": "Giskard Hub", "category": "Evaluation & testing", "description": "French vendor pairing an open-source Python testing library with a delivered assessment service. Automated and expert-led testing probes conversational agents for prompt injection, data disclosure, sycophancy, hallucination and inappropriate refusals, returning a severity-ranked vulnerability report and a signed go/no-go deployment recommendation.", "url": "https://www.giskard.ai/", "question": "Show me a full assessment report for my customer-facing agent, with vulnerabilities ranked by severity and the go/no-go deployment recommendation.", "reviewed": "2026-09-06" }, { "name": "GitGuardian NHI Governance", "category": "Agent identity & access", "description": "GitGuardian's module for non-human identity secrets: the ggscout collector inventories secrets and metadata from secrets managers, CI and infrastructure, tracks consumers, rotation dates and permissions, and flags stale or over-privileged credentials. It governs credentials used by machines and agents rather than issuing agent identities.", "url": "https://www.gitguardian.com/nhi-governance", "question": "Show me every secret outside our vaults, who consumes it, when it was last rotated, and the semi-automated rotation workflow for an over-privileged NHI secret.", "reviewed": "2026-09-06" }, { "name": "Google Cloud Agent Registry", "category": "Agent discovery & inventory", "description": "Centralized catalog in Gemini Enterprise Agent Platform for registering and governing AI agents, MCP servers, skills and endpoints. Agents can be registered automatically from supported runtimes or manually for custom deployments, and consumers can search the catalog and authenticate to registered tools. It governs registered assets; unregistered agents are out of scope.", "url": "https://docs.cloud.google.com/agent-registry/overview", "question": "Show me an ADK agent auto-registered from its runtime, then demonstrate searching the registry and binding it to an authenticated endpoint.", "reviewed": "2026-09-06" }, { "name": "Google Cloud Model Armor", "category": "Agent security & threat detection", "description": "Google Cloud service that screens LLM prompts and responses for prompt injection, jailbreaks, unsafe content and sensitive data, optionally returning sanitised text. Integrations extend screening to Google-managed MCP server traffic and the Gemini Enterprise agent platform, while the Agent Gateway integration is documented as preview.", "url": "https://docs.cloud.google.com/model-armor/overview", "question": "Show me Model Armor floor settings screening traffic to a Google-managed MCP server, blocking an injected prompt, and clarify which agent integrations are GA versus preview.", "reviewed": "2026-09-06" }, { "name": "GRACE Governance", "category": "AI risk & compliance management", "description": "Governance module of the Danish GRACE AI Platform: register AI projects and systems, define policies and controls, run AI assessments, report on control status and review compliance with audit trails and named role ownership. Sold as part of a wider platform, so standalone scope needs confirmation.", "url": "https://2021.ai/products/grace-governance", "question": "Show me how GRACE Governance is used without the rest of the GRACE platform, and what the control attestation and audit trail output looks like.", "reviewed": "2026-09-06" }, { "name": "Harmonic Security", "category": "Agent discovery & inventory", "description": "Browser-extension product that inventories employee AI usage by monitoring in-browser web traffic: which AI applications are used, whether the session uses a personal or corporate account, embedded AI features in sanctioned SaaS, and AI browsers. It also nudges or blocks sensitive prompt content. Coverage stops where the managed browser does.", "url": "https://www.harmonic.security/solutions/browser-based-genai-security", "question": "Show me a report distinguishing employees using ChatGPT on personal free accounts from those on our enterprise SSO tenant, and how the extension captured that distinction.", "reviewed": "2026-09-06" }, { "name": "Helicone", "category": "Observability & traceability", "description": "Apache-2.0 gateway and observability service that proxies LLM requests and logs them for inspection, with model routing across many providers. Capture is request/response oriented rather than agent-span oriented, and the pages reviewed document no RBAC, retention or redaction controls despite claiming SOC 2 and GDPR compliance.", "url": "https://docs.helicone.ai/", "question": "Show me the logged request and response for a specific model call routed through the gateway, and what data I can export via the API.", "reviewed": "2026-09-06" }, { "name": "Holistic AI AI Audits", "category": "Advisory & assurance services", "description": "Independent AI audit engagements from Holistic AI covering bias, privacy, efficacy, robustness and explainability, plus regulation-specific assessments and a separate independent audit service for Digital Services Act due-diligence obligations. Audits produce reports and mitigation strategies; no accreditation or certification mark is claimed on the pages reviewed.", "url": "https://www.holisticai.com/ai-audits", "question": "Show me an anonymised AI audit report structure and the specific tests you ran for bias and robustness on a deployed decisioning model.", "reviewed": "2026-09-06" }, { "name": "Holistic AI Governance Platform", "category": "AI risk & compliance management", "description": "Enterprise platform that inventories AI systems, models, agents and pipelines, then rates each system across risk verticals such as bias, robustness, privacy and transparency on a traffic-light dashboard. Marketed for internal builds and procured AI. Risk ratings are vendor-defined scores, not audits or certifications.", "url": "https://www.holisticai.com/ai-governance-platform", "question": "Demonstrate how a procured third-party model gets an inherent-risk rating and what evidence sits behind the red-amber-green score.", "reviewed": "2026-09-06" }, { "name": "HoneyHive", "category": "Observability & traceability", "description": "Agent observability and evaluation platform aimed at enterprises running production agents. Distributed tracing captures every step including tool calls, prompts, retries, loops and handoffs between sub-agents across long trajectories, with step-by-step replay, online evals on live traffic and annotation queues that turn expert review into datasets.", "url": "https://docs.honeyhive.ai/introduction", "question": "Replay a multi-day agent trajectory step by step, showing every tool call and sub-agent handoff plus the evaluation scores attached to it.", "reviewed": "2026-09-06" }, { "name": "IAPP Artificial Intelligence Governance Professional (AIGP)", "category": "Advisory & assurance services", "description": "Professional certification and matching online training from the IAPP that tests competency in AI development concepts, AI law and policy, and responsible AI governance and risk management. Individuals sit an exam against a published body of knowledge; the credential covers individual competency, not any organisation's AI system conformity.", "url": "https://iapp.org/certify/aigp", "question": "Show me the current AIGP body of knowledge and exam blueprint and how the training maps to EU AI Act obligations we must operationalise.", "reviewed": "2026-09-06" }, { "name": "IBM watsonx.governance", "category": "AI risk & compliance management", "description": "IBM's AI governance offering combining a tracked model and prompt-template inventory (AI Factsheets) with monitors for fairness, drift, model health and generative-AI output risks. Suited to enterprises already on watsonx; evidence reviewed covers monitoring and inventory mechanics, not regulatory outcomes.", "url": "https://www.ibm.com/products/watsonx-governance", "question": "Demonstrate how a deployed model's factsheet, fairness and drift monitors, and alert thresholds appear in the shared model inventory for a risk reviewer.", "reviewed": "2026-09-06" }, { "name": "Idira Secure AI Agents", "category": "Agent identity & access", "description": "The agentic module of Idira, Palo Alto Networks' identity security platform built on acquired CyberArk technology. It scans SaaS, cloud and developer environments for active agents, enriches them with ownership and permission context, brokers task-scoped access through an agent identity broker and logs agent actions for audit.", "url": "https://www.paloaltonetworks.com/idira/agentic", "question": "Show me agents discovered across our SaaS and cloud estate with owner and permission context, then a task-scoped grant issued and revoked by the agent identity broker.", "reviewed": "2026-09-06" }, { "name": "Immuta", "category": "Data governance & privacy", "description": "Data access governance platform that centralizes classification, policy authoring and query-time enforcement across warehouses and lakehouses, and can onboard RAG indexes and storage platforms as governed data sources. Its AI-specific evidence is limited to that onboarding capability; retrieval-time enforcement inside AI applications is not documented.", "url": "https://www.immuta.com/product/data-security-ai/", "question": "Demonstrate onboarding a RAG index as an Immuta data source and enforcing a row-level policy so the AI application cannot retrieve restricted records.", "reviewed": "2026-09-06" }, { "name": "Infosys Topaz Responsible AI Suite", "category": "Advisory & assurance services", "description": "Infosys implementation and advisory suite of more than ten offerings arranged as Scan, Shield and Steer, including regulatory watchtower monitoring, maturity and risk assessments, a responsible AI audit offering, and a control centre for compliance telemetry. Delivered as consulting plus proprietary assets; component depth is not documented publicly.", "url": "https://www.infosys.com/services/data-ai-topaz/offerings/responsible-ai-overview.html", "question": "Demonstrate the Responsible AI Control Center on a live AI portfolio: which compliance signals it ingests and what a violation alert looks like end to end.", "reviewed": "2026-09-06" }, { "name": "Inspect AI", "category": "Evaluation & testing", "description": "MIT-licensed evaluation framework from the UK AI Security Institute for running model and agent evaluations. It executes dataset samples through solvers, supports tool calling, multi-turn dialog, multi-agent primitives and sandboxed execution in Docker or Kubernetes, and writes evaluation logs for analysis. No hosted service or access control.", "url": "https://inspect.aisi.org.uk/", "question": "Demonstrate running one of the pre-built agentic evaluations against my model in a sandboxed environment and show the resulting evaluation log.", "reviewed": "2026-09-06" }, { "name": "JDLA C認証 (AI Governance Core Certification)", "category": "Advisory & assurance services", "description": "Japanese third-party certification scheme run by the Japan Deep Learning Association that reviews an organisation's AI governance structures and operations at legal-entity level, valid two years, with JDLA-accredited consulting firms supporting readiness separately from the review. It certifies governance arrangements, not individual AI tools or systems.", "url": "https://www.jdla.org/ai-governance/certification/", "question": "Show me the C認証 review criteria and the evidence a Japanese subsidiary must submit to demonstrate AI inventory and risk assessment practices.", "reviewed": "2026-09-06" }, { "name": "Knostic", "category": "Data governance & privacy", "description": "Israeli vendor whose platform maps enterprise content into a policy-aware knowledge graph, tests what enterprise LLM assistants will reveal, and adjusts access so answers respect need-to-know. Evidence comes from vendor product pages rather than technical documentation, and enforcement examples centre on Microsoft 365 Copilot content.", "url": "https://www.knostic.ai/the-genai-knowledge-security-platform/knowledge-controls", "question": "Demonstrate how Knostic detects that a Copilot answer exposed salary data to a user without need-to-know, and show the control change it applies.", "reviewed": "2026-09-06" }, { "name": "Kong AI Gateway (with MCP Traffic Gateway)", "category": "Runtime authorization & controls", "description": "Gateway layer from Kong that proxies LLM provider traffic and remote MCP traffic, exposes existing APIs as MCP tools, and applies authentication, access control and rate limiting plugins to those calls. Documentation emphasises traffic governance; semantic guardrails and agent threat detection are not covered on the fetched pages.", "url": "https://developer.konghq.com/ai-gateway/", "question": "Show me an internal REST API exposed as an MCP tool through Kong, protected by the AI MCP OAuth2 plugin, with a rate limit and traffic log for each tool call.", "reviewed": "2026-09-06" }, { "name": "KPMG AI Trust services", "category": "Advisory & assurance services", "description": "KPMG's multi-disciplinary AI governance service suite built on its Trusted AI framework, spanning AI risk assessment, AI systems inventory, governance and policy implementation, AI security and privacy, system cards, and an AI Assurance line offering model validation and independent attestation against defined frameworks such as SOC and HITRUST.", "url": "https://kpmg.com/xx/en/what-we-do/services/kpmg-trusted-ai/ai-trust-services.html", "question": "Show me an AI assurance and attestation engagement scope for an agentic workflow, including which framework you attest against and what the resulting report covers.", "reviewed": "2026-09-06" }, { "name": "Laminar", "category": "Observability & traceability", "description": "Apache-2.0 OpenTelemetry-native platform for AI agents that traces LLM calls, tool use, custom functions and parallel sub-agents, presenting runs as transcripts rather than span trees. It records and replays runs in a debugger, builds evaluation datasets from production traces, and runs evals in CI. No documented access control.", "url": "https://docs.lmnr.ai/", "question": "Replay a recorded agent run in the debugger, swap the prompt, and show the side-by-side comparison plus the eval that would catch the regression in CI.", "reviewed": "2026-09-06" }, { "name": "Langfuse", "category": "Observability & traceability", "description": "German open-source AI engineering platform whose traces record LLM and non-LLM calls, retrieval, embeddings and API calls, grouped into sessions for multi-turn conversations and agent graphs. Paid tiers add audit logs, fine-grained RBAC, SSO/SCIM, retention management, client-side masking and EU, US or JP data regions.", "url": "https://langfuse.com/docs", "question": "Show me a session trace for a multi-turn agent conversation, then show the audit log of who accessed or changed that project.", "reviewed": "2026-09-06" }, { "name": "LangSmith", "category": "Observability & traceability", "description": "LangChain's hosted platform for tracing LLM and agent applications. Traces record what agents did in production and feed evaluation datasets, online evaluation rules and annotation queues. Enterprise plans add role-based access control with workspace roles and custom organization roles; audit logging is not documented.", "url": "https://docs.smith.langchain.com/", "question": "Show me a production trace for a failed agent run and demonstrate which workspace roles can view or edit that trace.", "reviewed": "2026-09-06" }, { "name": "LangSmith Deployment (LangGraph)", "category": "Agent building & orchestration", "description": "LangChain's runtime for running agents in production alongside the open-source LangGraph library. It provides durable execution, threads and runs, and middleware that pauses tool calls for human approval, edit or rejection. Governance beyond approvals and tracing is developer-configured rather than an administrative control surface.", "url": "https://docs.langchain.com/langgraph-platform/index", "question": "Show me an agent deployed on LangSmith Deployment pausing on a database-write tool call, and a reviewer editing the arguments before it resumes.", "reviewed": "2026-09-06" }, { "name": "LangWatch", "category": "Observability & traceability", "description": "Dutch open-source LLMOps platform that automatically tracks every LLM call, tool usage and user interaction as traces, spans and metadata, adds prompt versioning, dataset experiments and multi-turn agent testing against a simulated user. Self-hosted and enterprise editions document audit logs, RBAC and SSO/SCIM.", "url": "https://docs.langwatch.ai/", "question": "Show me the full conversation trace for one agent session, then show the audit log entry for the user who changed the evaluation configuration.", "reviewed": "2026-09-06" }, { "name": "LlamaFirewall", "category": "Agent security & threat detection", "description": "Open-source guardrail framework from Meta that runs layered scanners around agent execution: PromptGuard 2 for jailbreak detection, Agent Alignment Checks that audit chain-of-thought for goal misalignment, and CodeShield static analysis of generated code. Meta describes the alignment auditor as still experimental.", "url": "https://meta-llama.github.io/PurpleLlama/LlamaFirewall/", "question": "Show me LlamaFirewall scanning an email agent's inputs with PromptGuard 2 and flagging an indirect injection through Agent Alignment Checks, including the experimental caveats.", "reviewed": "2026-09-06" }, { "name": "Lumenova AI Platform", "category": "AI risk & compliance management", "description": "AI governance platform aimed at regulated industries that replaces manual risk reviews with workflow-driven assessments, evaluates models across a large metric set, and adds monitoring plus guardrails. Public evidence is largely vendor-authored, including a 2024 UK government assurance-technique listing.", "url": "https://www.lumenova.ai/platform/", "question": "Show me which of the 200+ model metrics are computed automatically for a deployed model and how results feed the risk assessment workflow.", "reviewed": "2026-09-06" }, { "name": "Maxim AI", "category": "Evaluation & testing", "description": "India-founded platform combining agent simulation, evaluation and production logging. Teams simulate multi-turn interactions across scenarios and personas, score runs with AI, programmatic or statistical evaluators plus human review pipelines, curate datasets from real interactions, and analyse production logs with distributed tracing.", "url": "https://www.getmaxim.ai/docs", "question": "Show me a simulation run across several user personas with evaluator scores per version, and the production logs for the same agent.", "reviewed": "2026-09-06" }, { "name": "Microsoft Agent 365", "category": "Agent discovery & inventory", "description": "Admin control plane for AI agents in Microsoft 365. Its registry lists Microsoft, partner-built, org-published and user-shared agents available to a tenant, and a preview Shadow AI view detects unapproved local agents on managed devices using Defender for Endpoint, with extra usage metadata from Global Secure Access. Registry coverage is Microsoft-ecosystem centric.", "url": "https://learn.microsoft.com/en-us/microsoft-365/admin/manage/agent-registry?view=o365-worldwide", "question": "Demonstrate the Shadow AI page detecting an unapproved local agent on a Defender-enrolled device and show which governance action removes its access.", "reviewed": "2026-09-06" }, { "name": "Microsoft Entra Agent ID", "category": "Agent identity & access", "description": "Microsoft's extension of Entra ID that creates dedicated directory identities for AI agents, built from reusable agent identity blueprints, so agents authenticate with their own credentials and are governed by Conditional Access, lifecycle and cleanup workflows. Evidence covers Microsoft-centric estates; cross-vendor agent coverage is not established.", "url": "https://learn.microsoft.com/en-us/entra/agent-id/what-are-agent-identities", "question": "Show me how you create an agent identity from a blueprint, bind it to a sponsor and owner, and enforce Conditional Access on that agent's token request.", "reviewed": "2026-09-06" }, { "name": "Microsoft Foundry Agent Service", "category": "Agent building & orchestration", "description": "Managed Microsoft platform for building, deploying and scaling agents, offering prompt agents configured in a portal, hosted container agents from frameworks such as Agent Framework and LangGraph, and direct Responses API use. Each hosted agent receives a dedicated Microsoft Entra identity plus managed endpoint, scaling and observability.", "url": "https://learn.microsoft.com/en-us/azure/foundry/agents/overview", "question": "Show me a hosted agent deployed from a container that receives its own Entra identity, and the end-to-end trace of one of its runs.", "reviewed": "2026-09-06" }, { "name": "Microsoft Purview Data Security Posture Management for AI", "category": "Data governance & privacy", "description": "Microsoft's Purview module that reports on how organizational data is used by Copilot experiences, agents and third-party AI sites, surfaces oversharing risk, and applies ready-made data-loss policies to AI prompts. Coverage of non-Microsoft AI sites depends on device onboarding and a browser extension, so unmanaged endpoints stay invisible.", "url": "https://learn.microsoft.com/en-us/purview/dspm-for-ai", "question": "Show me the Apps and agents dashboard listing every agent in my tenant, the sensitive data each one accessed, and which Purview policy protected it.", "reviewed": "2026-09-06" }, { "name": "Mindgard", "category": "Evaluation & testing", "description": "UK vendor running automated red-team tests against AI models, applications and agents. It profiles the target, enumerates attack surface, executes techniques from a maintained attack library via CLI or SDK, and reports exploitable findings with remediation guidance. Documentation covers testing mechanics rather than audit, retention or access controls.", "url": "https://mindgard.ai/", "question": "Show me an automated red-team run against my production agent, including which attack techniques were executed and the remediation guidance produced.", "reviewed": "2026-09-06" }, { "name": "MintMCP", "category": "Runtime authorization & controls", "description": "Managed MCP gateway for enterprises that curates which MCP servers and tools each role may reach, issues per-agent credentials, and logs every tool call. An agent monitor captures file reads, command execution and tool calls so teams can write rules that block risky behaviour. Self-hosting requires vendor contact.", "url": "https://www.mintmcp.com/", "question": "Show me two role-based MintMCP endpoints where a destructive tool is switched off for one role, plus the agent monitor rule that blocks a risky command.", "reviewed": "2026-09-06" }, { "name": "MLflow Model Registry", "category": "Model lifecycle & governance", "description": "Open-source component of MLflow providing a central model store with named registered models, versions, aliases, tags and lineage back to the producing run, used by ML teams as the promotion and rollback backbone. It provides mechanics for staging, not policy, risk assessment or regulatory mapping.", "url": "https://mlflow.org/docs/latest/ml/model-registry/", "question": "Demonstrate promoting a model version via alias with lineage back to its run, and show what audit information the registry retains after rollback.", "reviewed": "2026-09-06" }, { "name": "ModelOp Center", "category": "Model lifecycle & governance", "description": "Model and AI lifecycle governance software that maintains a searchable inventory of ML models, generative AI, agents, vendor tools and embedded SaaS AI, routes intake through policy-driven workflows, and maps controls to regulations including SR 11-7, the EU AI Act, NIST AI RMF and ISO 42001.", "url": "https://www.modelop.com/ai-governance-software/inventory", "question": "Show me a policy-driven workflow blocking a non-compliant model promotion, with the control mapping and captured sign-off evidence.", "reviewed": "2026-09-06" }, { "name": "Modulos AI Governance Platform", "category": "AI risk & compliance management", "description": "Governance platform organised around projects representing one AI system each, with frameworks, requirements, controls, evidence and reviews, plus a risk module that quantifies AI risk in monetary terms and a runtime inspection module for scheduled tests. Multi-framework mapping is vendor-maintained content, not a compliance guarantee.", "url": "https://docs.modulos.ai/platform/overview", "question": "Demonstrate how one control satisfies EU AI Act and ISO 42001 requirements simultaneously and how the linked evidence and review trail is exported.", "reviewed": "2026-09-06" }, { "name": "n8n", "category": "Agent building & orchestration", "description": "German workflow automation tool, source-available and self-hostable, whose AI Agent node connects a chat model to tools so the agent chooses which to call. Governance is workspace-level rather than agent-level: role-based access control across instance and project scopes, with no documented per-action approval or agent audit trail.", "url": "https://docs.n8n.io/integrations/builtin/cluster-nodes/root-nodes/n8n-nodes-langchain.agent/", "question": "Show me an AI Agent workflow in a project where a Member role can execute but not edit it, and explain what record exists of each tool the agent called.", "reviewed": "2026-09-06" }, { "name": "Naaia", "category": "AI risk & compliance management", "description": "French platform marketed as an AI management system aligned to ISO/IEC 42001 that qualifies an organisation's operator status and system risk level under the EU AI Act, then generates an operational action plan with a centralised registry of projects, systems, models and components.", "url": "https://naaia.ai/en/eu-ai-act/", "question": "Show me how operator status and EU AI Act risk level are qualified for one system and what the generated action plan and documentation pack contain.", "reviewed": "2026-09-06" }, { "name": "NannyML", "category": "Model lifecycle & governance", "description": "Open-source Python library for post-deployment monitoring that estimates a model's performance when ground-truth labels are delayed or missing, using confidence-based estimation for classification and direct loss estimation for regression, and links univariate and multivariate drift alerts to performance impact.", "url": "https://github.com/NannyML/NannyML", "question": "Demonstrate estimated versus realised ROC AUC on a tabular classifier with delayed labels, and show which drift alerts were linked to the performance change.", "reviewed": "2026-09-06" }, { "name": "Netskope One AI Command Center", "category": "Agent discovery & inventory", "description": "Module of Netskope One AI Security that discovers AI assets - corporate or personal, managed or shadow, cloud or on-premises - from the vendor's SSE/proxy vantage point and maps them to the identities, data stores and tools they connect to, adding risk correlation and response. Discovery leans on traffic and platform telemetry rather than code scanning.", "url": "https://www.netskope.com/solutions/netskope-one-ai-security", "question": "Show me the asset-to-identity-to-data-store map for a shadow AI application discovered from our traffic, including any MCP servers it reaches.", "reviewed": "2026-09-06" }, { "name": "NeuralTrust (TrustGate)", "category": "Agent security & threat detection", "description": "Barcelona-based platform combining an agent gateway (TrustGate) with runtime protection over the models, tools, MCP servers and data agents touch. Documented gateway behaviour includes per-user and per-tool RBAC, end-user identity forwarding across hops and cryptographic audit trails, with SaaS, hybrid and air-gapped deployment options.", "url": "https://neuraltrust.ai/", "question": "Show me TrustGate forwarding end-user identity through two agent hops while denying a payments API tool for that user, plus the cryptographic audit record of each tool call.", "reviewed": "2026-09-06" }, { "name": "Noma Security Platform", "category": "Agent security & threat detection", "description": "Platform that inventories agents, MCP servers, skills and models across endpoints, SaaS agent builders and homegrown AI stacks, maps each agent's permissions and data access, red teams them before production, and evaluates runtime actions to alert, block, mask data or route to a human. Claims rest on vendor pages.", "url": "https://noma.security/platform", "question": "Show me Noma discovering an unapproved MCP server on a developer laptop, mapping its blast radius, and then routing a risky agent action to a human for approval.", "reviewed": "2026-09-06" }, { "name": "Nudge Security AI Agent Discovery", "category": "Agent discovery & inventory", "description": "Research-preview capability of Nudge Security that inventories AI agents employees create on platforms such as Copilot Studio, Agentforce, Gemini, OpenAI, n8n, ServiceNow, Databricks, Workato and Tines, surfacing creator, permissions, connected data and MCP connections. Nudge's discovery approach centres on corporate email signals and OAuth grants.", "url": "https://www.nudgesecurity.com/try-nudge/ai-agent-discovery", "question": "Demonstrate discovering an n8n or Copilot Studio agent an employee built last week, showing the creator, its OAuth grants, and any unauthenticated MCP connection.", "reviewed": "2026-09-06" }, { "name": "NVIDIA NeMo Guardrails", "category": "Agent security & threat detection", "description": "Open-source Python toolkit that intercepts LLM application inputs, outputs and custom action calls, applying configurable rails written in YAML and Colang to block or modify content. The guardrail catalogue includes jailbreak detection, PII handling and agentic security checks; the repository labels the release beta and not production-recommended.", "url": "https://github.com/NVIDIA/NeMo-Guardrails/", "question": "Show me NeMo Guardrails applying an execution rail around a tool-calling action plus jailbreak detection, and explain the beta production caveat in the repository.", "reviewed": "2026-09-06" }, { "name": "Oasis Agentic Access Management (AAM)", "category": "Agent identity & access", "description": "Oasis places an access layer between AI agents and SaaS, cloud, on-prem and data systems: it converts each agent request into a structured intent, evaluates it against policy with optional human escalation, and provisions ephemeral least-privilege session identities with prompt-level audit trails. Enforcement depends on agents routing through Oasis.", "url": "https://www.oasis.security/blog/introducing-oasis-agentic-access-management", "question": "Demonstrate an agent request converted into structured intent, denied by policy, then approved with human-in-the-loop and executed under a just-in-time session identity.", "reviewed": "2026-09-06" }, { "name": "Obsidian Security Shadow AI", "category": "Agent discovery & inventory", "description": "Module of Obsidian's SaaS security platform that builds a continuously updated inventory of AI tools and agents by combining a managed browser extension, API integrations into SaaS tenants, and mapping of agent-to-MCP connections. Aimed at security teams; agent coverage depends on which SaaS tenants and endpoints are instrumented.", "url": "https://www.obsidiansecurity.com/shadow-ai-security", "question": "Show me an agent discovered only by your browser extension that never appeared in the SaaS platform's own API-reported agent list, with its creator, permissions, and MCP connections.", "reviewed": "2026-09-06" }, { "name": "Okta for AI Agents", "category": "Agent identity & access", "description": "Okta's agent-focused offering that discovers known and shadow AI agents, registers them as identities in Universal Directory with a human owner, and brokers short-lived credentials plus secret vaulting for agent access. Discovery evidence in its posture-management documentation is limited to specific connected agent platforms.", "url": "https://www.okta.com/products/govern-ai-agent-identity/", "question": "Show me an unmanaged Salesforce Agentforce agent discovered via OAuth consent grants, then register it in Universal Directory with an owner and issue it a short-lived credential.", "reviewed": "2026-09-06" }, { "name": "OneTrust AI Governance", "category": "AI risk & compliance management", "description": "AI-specific module of OneTrust's privacy and governance suite: central inventory of AI systems, models, datasets, agents and vendors with use-case intake and approval workflows, risk tiering, impact assessments and policy-driven controls intended to produce audit-ready records for security and governance teams.", "url": "https://www.onetrust.com/products/ai-governance/", "question": "Show me the AI use-case intake and approval workflow end to end, and which evidence OneTrust captures automatically versus manually.", "reviewed": "2026-09-06" }, { "name": "OpenFGA", "category": "Runtime authorization & controls", "description": "Apache-2.0 relationship-based authorization engine whose documentation models agents as first-class principals with narrowly scoped, revocable, optionally time-limited grants, including MCP tool-level checks and permission-filtered retrieval. It answers authorization questions but does not intercept traffic, so an application or gateway must call it.", "url": "https://openfga.dev/docs/use-cases/ai-agent-authorization", "question": "Show me an OpenFGA model where an agent receives a task-scoped grant to two MCP tools with a turn limit, then revoke the agent without touching the user's permissions.", "reviewed": "2026-09-06" }, { "name": "OpenLLMetry", "category": "Observability & traceability", "description": "Apache-2.0 set of OpenTelemetry extensions from Traceloop that instrument GenAI applications non-intrusively in Python, TypeScript or Go and export traces either to Traceloop or to an existing observability backend. It is instrumentation plumbing: no access control, retention or evaluation features are part of the library.", "url": "https://www.traceloop.com/docs/openllmetry/introduction", "question": "Demonstrate exporting GenAI traces from my agent to my own OpenTelemetry backend, showing the workflow, task, agent and tool annotations.", "reviewed": "2026-09-06" }, { "name": "OpenSSF Model Signing (OMS) / model-transparency", "category": "Model lifecycle & governance", "description": "OpenSSF-backed specification with an Apache-2.0 library and CLI that signs and verifies machine learning model artifacts using Sigstore, self-signed certificates, public keys or PKCS#11 devices, producing signature bundles that let consumers check model integrity and provenance before deployment or reuse.", "url": "https://openssf.org/projects/model-signing/", "question": "Demonstrate signing a multi-gigabyte model with Sigstore and verifying the signature in a deployment pipeline gate, including what the bundle attests to.", "reviewed": "2026-09-06" }, { "name": "Operant Semantic Firewall", "category": "Runtime authorization & controls", "description": "Inline enforcement layer that inspects prompts, plans, tool calls, generated commands and data payloads before execution and returns an allow, block or redact decision, with a companion MCP gateway applying least-privilege controls and trust zones. Evidence comes from vendor product pages rather than reference documentation.", "url": "https://www.operant.ai/platform/semantic-firewall", "question": "Show me the Semantic Firewall blocking a shell command produced by an injected instruction and redacting a bulk data read, with the policy that produced each decision.", "reviewed": "2026-09-06" }, { "name": "Opik", "category": "Observability & traceability", "description": "Apache-2.0 platform from Comet that records every LLM call, tool invocation and agent step as traces, monitors cost, latency, error rates and feedback scores, scores incoming traces with online rules, and converts problem traces into test cases. The full backend can be self-hosted; RBAC and audit logs are not documented.", "url": "https://www.comet.com/docs/opik/", "question": "Show me a self-hosted deployment where every tool invocation in an agent run is logged and automatically scored by an online evaluation rule.", "reviewed": "2026-09-06" }, { "name": "ORCAA Algorithmic Audit", "category": "Advisory & assurance services", "description": "ORCAA sells algorithmic audits that assess risks of a specific algorithmic use case using its Ethical Matrix framework, plus quantitative bias testing for regulatory compliance such as New York City Local Law 144 bias audits, AI governance consultation, and metric 'cockpit' design. Deliverables are reports, not certifications.", "url": "https://orcaarisk.com/", "question": "Show me how the Ethical Matrix is populated for one hiring algorithm and what a Local Law 144 bias audit report you issued contains.", "reviewed": "2026-09-06" }, { "name": "Orkes Conductor", "category": "Agent building & orchestration", "description": "Commercial platform built on the Conductor durable orchestration engine. Conductor Agents are authored in code or imported from frameworks, compiled into versioned workflow definitions, and composed with branching, schedules, human approval and cancellation, with each LLM call and tool invocation visible in the UI and API.", "url": "https://orkes.io/content/devguide/ai/conductor-agents", "question": "Show me a Conductor Agent deployed at a pinned version, invoked from a production workflow, with each LLM call and tool invocation visible in the execution view.", "reviewed": "2026-09-06" }, { "name": "Palo Alto Networks AI Access Security", "category": "Agent discovery & inventory", "description": "SASE-delivered product that identifies which generative AI applications employees are using by matching network traffic against a maintained dictionary of GenAI apps grouped into predefined use cases, then applying access-control and DLP policy. It discovers app usage, not internally built agents or AI components in code.", "url": "https://docs.paloaltonetworks.com/ai-access-security", "question": "Demonstrate the Insights dashboard showing the top GenAI apps by user and use case in our traffic, and how a new app appears once it is added to the app dictionary.", "reviewed": "2026-09-06" }, { "name": "Patronus AI", "category": "Evaluation & testing", "description": "Vendor offering managed evaluators plus simulation environments for agent testing: hosted judges score hallucination and unsafe output, red-teaming algorithms probe for weaknesses, and simulated digital workflows exercise long-horizon agent tasks. Public pages document scoring and simulation but not audit records, retention or access control.", "url": "https://docs.patronus.ai/", "question": "Demonstrate a simulated multi-step workflow run where my agent is scored for hallucination and unsafe output, and show what evaluation evidence I can export.", "reviewed": "2026-09-06" }, { "name": "Permit MCP Gateway", "category": "Runtime authorization & controls", "description": "Enforcement proxy placed between MCP clients such as Cursor or Claude Desktop and upstream MCP servers. It authenticates the human behind an agent, checks each tool call against fine-grained policy, records allow and deny decisions, and requires no change to existing MCP servers. Policy authoring depends on Permit's control plane.", "url": "https://www.permit.io/mcp-gateway", "question": "Show me a Cursor session where a write-capable MCP tool is denied by policy while a read tool succeeds, then show the audit entry naming the agent and the authorizing human.", "reviewed": "2026-09-06" }, { "name": "Pillar Security", "category": "Agent security & threat detection", "description": "Israeli platform covering the AI agent lifecycle: cataloguing agents, models, prompts, MCP servers and coding agents through agentless integrations, then applying runtime guardrails that monitor prompts, tool calls and commands for prompt injection, tool poisoning and data exfiltration. Product claims come from vendor pages, not reference docs.", "url": "https://www.pillar.security/", "question": "Show me Pillar validating that an agent's tool call matches its declared schema, flagging a permission-scope deviation, and blocking a poisoned instruction in an agent-to-agent handoff.", "reviewed": "2026-09-06" }, { "name": "Planview Agent Resource Management", "category": "Work discovery & human delegation", "description": "Extension of Planview's portfolio resource model that treats AI agents as resources alongside people, so leaders can plan and allocate blended human and agent capacity across strategic work, track associated cost, and tie each agent action back to an accountable human decision-maker. It is planning and accountability tooling, not an agent runtime.", "url": "https://newsroom.planview.com/planview-launches-agent-resource-management-redefining-portfolio-resource-management-for-the-ai-era/", "question": "Show me a portfolio view where agent and human capacity are planned together, with the cost of agent work and the named human accountable for each agent's assignments.", "reviewed": "2026-09-06" }, { "name": "Presidio", "category": "Data governance & privacy", "description": "MIT-licensed SDK, originally from Microsoft and now transitioning to community ownership, that detects PII in text, images and structured data using recognizers, regex, rules and checksums, then anonymizes it with configurable operators. It is a developer library with no policy console, access control, lineage or audit features.", "url": "https://github.com/microsoft/presidio", "question": "Show me a custom recognizer added to Presidio that detects our internal customer ID format and anonymizes it in a prompt before an LLM call.", "reviewed": "2026-09-06" }, { "name": "Prisma AIRS AI Runtime Security", "category": "Agent security & threat detection", "description": "Palo Alto Networks security service that scans prompts and model responses via API or network enforcement to detect prompt injection, sensitive data leakage and malicious content, with agent-focused detections such as MCP threat detection, tool chaining attack analysis and privilege misuse. Delivered as a managed enterprise service.", "url": "https://docs.paloaltonetworks.com/ai-runtime-security", "question": "Show me Prisma AIRS scanning an agent request through the AI Runtime API, flagging a tool chaining attack and a prompt injection, and the security profile that blocked it.", "reviewed": "2026-09-06" }, { "name": "Private AI", "category": "Data governance & privacy", "description": "Canadian vendor providing a container-deployable API that identifies and removes personal identifiers from text before it reaches a model, with optional synthetic replacement and consistent markers. It is a detection and de-identification layer only: no access control, lineage, consent or audit features appear in its API reference.", "url": "https://docs.private-ai.com/reference/2.9/overview/", "question": "Demonstrate the de-identify call redacting 20 PII classes from a prompt and then re-identifying the response, with entity relationships preserved.", "reviewed": "2026-09-06" }, { "name": "Project Moonshot", "category": "Evaluation & testing", "description": "Apache-2.0 LLM evaluation toolkit from Singapore's AI Verify Foundation that combines benchmark testing across safety and performance metrics with manual and automated red-teaming, offers guided workflows for IMDA's starter kit for LLM app testing, and produces shareable scoring reports usable in CI pipelines.", "url": "https://github.com/aiverify-foundation/moonshot", "question": "Show me running IMDA's starter-kit benchmarks plus an automated red-team attack module against our chatbot and the scoring report it produces.", "reviewed": "2026-09-06" }, { "name": "Promptfoo", "category": "Evaluation & testing", "description": "Open-source evaluation and red-teaming tool that generates adversarial inputs from configurable plugins, runs them against an LLM application, and grades outputs with deterministic and model-graded metrics in CI. The paid enterprise editions add RBAC and team scoping; audit logging and retention are not documented.", "url": "https://www.promptfoo.dev/docs/red-team/", "question": "Demonstrate an end-to-end red-team scan of my deployed agent, including which plugins ran and how findings are scoped to a team.", "reviewed": "2026-09-06" }, { "name": "Protecto", "category": "Data governance & privacy", "description": "India- and US-based vendor offering deterministic tokenization for sensitive values used in LLM prompts and AI pipelines, with policy-controlled unmasking that requires an explicit request and sufficient caller permissions. Documentation covers tokenization mechanics; discovery, lineage, consent and audit logging are not described.", "url": "https://docs.protecto.ai/core-concepts/tokenization-basics", "question": "Demonstrate masking a support ticket before it goes to an LLM, then an unmask request being denied because the active policy does not permit it.", "reviewed": "2026-09-06" }, { "name": "PwC Responsible AI Toolkit", "category": "Advisory & assurance services", "description": "PwC advisory offering delivered as customisable frameworks, tools and processes for enterprise AI governance: role and responsibility design across three lines of defence, regulatory monitoring, policy development, bias and fairness assessment, and a free entry-level Responsible AI Diagnostic. Pages do not disclose tooling detail or pricing.", "url": "https://www.pwc.com/sg/en/services/reimagine-digital/data-optimisation/what-is-responsible-ai.html", "question": "Show me how the Responsible AI Diagnostic output maps to a concrete governance remediation roadmap with three-lines-of-defence ownership.", "reviewed": "2026-09-06" }, { "name": "PyRIT", "category": "Evaluation & testing", "description": "MIT-licensed Python framework from Microsoft for probing generative AI systems for risk. It is aimed at security engineers running automated adversarial testing campaigns rather than at governance teams, and the repository provides no multi-user controls, evidence retention or reporting workflow of its own.", "url": "https://github.com/microsoft/PyRIT", "question": "Demonstrate an automated PyRIT attack run against my deployed model endpoint and show what artefacts the run leaves behind.", "reviewed": "2026-09-06" }, { "name": "Ragas", "category": "Evaluation & testing", "description": "Open-source Python evaluation library for LLM and RAG applications. It runs LLM-based and deterministic metrics over datasets, generates synthetic test sets, and tracks results across experiments so changes can be compared. It is a local library with no hosted control plane, access control or audit trail.", "url": "https://docs.ragas.io/en/stable/", "question": "Show me a Ragas experiment comparing two retrieval configurations on a generated test set, with per-metric scores and reasons.", "reviewed": "2026-09-06" }, { "name": "Resaro Approved Intelligence", "category": "Advisory & assurance services", "description": "Independent AI assurance firm running testing, evaluation, validation and verification workflows in the client's own environment via its Approved Intelligence Platform, producing structured deployment evidence and continuous post-deployment evaluation for civil and defence uses. Marketed as evidence for deployment decisions rather than as certification against a standard.", "url": "https://resaro.ai/", "question": "Demonstrate a TEVV run in our environment for one mission-critical model and show the evidence pack a deployment authority receives.", "reviewed": "2026-09-06" }, { "name": "Saidot", "category": "AI risk & compliance management", "description": "Finnish AI governance platform that links registered systems, models, agents and datasets to an expert-curated graph of risks, controls and policy requirements, so recommendations and risk inheritance propagate to connected assets. Curated content speeds setup but reflects the vendor's interpretation of obligations.", "url": "https://www.saidot.ai/product", "question": "Demonstrate risk inheritance: register a system on a new foundation model and show which risks, controls and policy requirements are auto-recommended and why.", "reviewed": "2026-09-06" }, { "name": "SailPoint Agent Identity Security", "category": "Agent identity & access", "description": "Part of SailPoint Identity Security Cloud, this module aggregates AI agents from AWS, Azure, GCP, Salesforce and Copilot Studio, onboards each as a registered identity with business and access context, and surfaces shadow AI usage. It complements SailPoint's separate machine identity module for service accounts, bots and RPAs.", "url": "https://www.sailpoint.com/products/agent-identity-security", "question": "Show me every AI agent aggregated from Copilot Studio and Salesforce, each registered with an owner and access context, plus the shadow AI usage report.", "reviewed": "2026-09-06" }, { "name": "Securiti Gencore AI", "category": "Data governance & privacy", "description": "Securiti's pipeline product for building enterprise AI systems on governed data. It catalogs unstructured sources, redacts or masks sensitive values inside AI pipelines, loads permission-aware embeddings into vector databases, and tracks lineage from source file to embedding. Detailed technical documentation was not publicly reachable during review.", "url": "https://securiti.ai/gencore/", "question": "Show me a Gencore AI pipeline where a retrieval answer changes because the requesting user lacks entitlements on the underlying SharePoint file.", "reviewed": "2026-09-06" }, { "name": "ServiceNow AI Control Tower", "category": "AI risk & compliance management", "description": "ServiceNow application that inventories AI agents, models and MCP servers as configuration items tied to the CMDB, with persona-based views for AI stewards, owners and risk/compliance users, and lifecycle plus risk and compliance oversight for platform customers already using ServiceNow.", "url": "https://www.servicenow.com/products/ai-control-tower.html", "question": "Demonstrate how a discovered third-party AI agent becomes a CMDB configuration item with owner, lineage and a risk assessment task for the AI steward.", "reviewed": "2026-09-06" }, { "name": "Skyflow LLM Privacy Vault", "category": "Data governance & privacy", "description": "Skyflow's pattern for isolating sensitive values in a data privacy vault so prompts, training sets and RAG content carry tokens instead of raw identifiers, with role-based re-identification for authorized processes. It governs sensitive fields rather than discovering data estate-wide, and requires application changes to adopt.", "url": "https://docs.skyflow.com/docs/fundamentals/patterns/llm-privacy", "question": "Show me a RAG flow where the vault de-identifies customer records before embedding and only a privileged role can re-identify the values in the answer.", "reviewed": "2026-09-06" }, { "name": "Snyk Agent Scan (formerly MCP-Scan)", "category": "Agent security & threat detection", "description": "Apache-2.0 command line scanner that discovers locally installed agent components — harnesses, MCP servers, skills — and checks tools, prompts and resources for prompt injection, tool poisoning, cross-origin escalation and tool changes, with a proxy mode that inspects live MCP traffic. Some checks call Snyk's hosted API.", "url": "https://github.com/snyk/agent-scan", "question": "Demonstrate scanning our developers' MCP configurations and show what a tool-poisoning and rug-pull finding looks like, plus which checks require the hosted API.", "reviewed": "2026-09-06" }, { "name": "SPIRE (SPIFFE Runtime Environment)", "category": "Agent identity & access", "description": "A CNCF graduated open-source toolchain that attests running workloads and issues them short-lived SPIFFE identities (X.509 and JWT SVIDs) through the Workload API, enabling mutual TLS and authentication to secret stores, databases and cloud services. It provides identity plumbing, not agent inventory or governance workflows.", "url": "https://github.com/spiffe/spire", "question": "Demonstrate a Kubernetes workload attesting to SPIRE, receiving a rotating X.509 SVID, and using it for mTLS to another service across a federated trust domain.", "reviewed": "2026-09-06" }, { "name": "SPLX AI Asset Management", "category": "Agent discovery & inventory", "description": "AI-BOM and inventory module of the SPLX platform, now part of Zscaler. It connects to cloud platforms, code repositories and ML/AI platforms to detect LLMs in use, scan repositories to map agents, tools and MCP servers in AI workflows, and run risk assessments on discovered agents. Detection is scan-based, not user-traffic based.", "url": "https://splx.ai/platform/ai-asset-management", "question": "Show me the agentic workflow map produced from scanning one of our repositories, including each agent, its tools, and the MCP servers it connects to.", "reviewed": "2026-09-06" }, { "name": "Straiker Defend AI", "category": "Agent security & threat detection", "description": "Runtime security product for AI agents that inspects prompts, reasoning steps and tool calls across coding assistants, productivity copilots and custom agents, blocking direct and indirect injection, destructive actions such as file deletion, and data exfiltration. Vendor pages also describe shutting down rogue agents and connections.", "url": "https://www.straiker.ai/products/defend-ai", "question": "Show me Defend AI blocking an indirect injection delivered in an email to a productivity copilot and stopping a coding agent from deleting files, then show the rogue-agent shutdown action.", "reviewed": "2026-09-06" }, { "name": "Teleport Machine & Workload Identity", "category": "Agent identity & access", "description": "Teleport issues short-lived certificates, JWTs and SPIFFE-compatible identities to bots, CI jobs, workloads and AI agents via its tbot agent, then enforces role-based access and records every command an agent runs against SSH hosts, Kubernetes clusters, databases and MCP servers. Agent use cases are positioned around design partnerships.", "url": "https://goteleport.com/docs/machine-workload-identity/use-cases/ai-agents-mwi/", "question": "Show an AI agent issued its own Teleport identity, restricted to read-only Kubernetes access by RBAC, with the full session recording of its commands.", "reviewed": "2026-09-06" }, { "name": "Temporal", "category": "Agent building & orchestration", "description": "Durable execution platform used to run stateful agent loops that call models and tools, wait on human decisions for days, and resume exactly where they stopped after failures. It supplies reliability primitives and full execution history rather than AI-specific policy, review consoles or data controls.", "url": "https://docs.temporal.io/ai", "question": "Demonstrate an agent workflow that pauses for multi-day human approval via a Signal, survives a worker crash, and replays its full execution history.", "reviewed": "2026-09-06" }, { "name": "Token Security NHI & AI Agent Security Platform", "category": "Agent identity & access", "description": "Token Security continuously discovers AI agents and non-human identities across on-prem, hybrid and cloud estates, correlates agents, humans, secrets, permissions and data in an identity graph, enforces ownership and governs agents from creation through retirement, decommissioning orphaned identities. Credential issuance is not evidenced.", "url": "https://www.token.security/", "question": "Show me a custom GPT agent discovered in our tenant, its owner and blast radius in the identity graph, and the workflow that decommissions it when the owner leaves.", "reviewed": "2026-09-06" }, { "name": "Tonic Textual", "category": "Data governance & privacy", "description": "Tonic.ai's de-identification product for unstructured text. It scans files and strings with named-entity models, then redacts or replaces sensitive values with consistent synthetic substitutes for use in LLM prompts, RAG corpora and training sets, with guided review. Governance features beyond RBAC and SSO are not documented.", "url": "https://docs.tonic.ai/textual", "question": "Show me a guided redaction project that de-identifies a folder of PDFs, then the SDK call that redacts a live prompt with the same entity configuration.", "reviewed": "2026-09-06" }, { "name": "trail", "category": "AI risk & compliance management", "description": "German AI governance tool centred on a live AI registry of use cases, systems, agents, models and third-party vendors, with guided EU AI Act risk and role classification, curated framework templates, control implementation tracking and reusable assessments and evidence across linked assets.", "url": "https://www.trail-ml.com/ai-governance", "question": "Show me how EU AI Act role and risk classification is derived for one system and how a completed assessment is reused on a dependent asset.", "reviewed": "2026-09-06" }, { "name": "Trustible AI Governance Platform", "category": "AI risk & compliance management", "description": "Purpose-built system of record for AI intake: teams capture use cases, models, agents and vendors, get risk-based triage, run risk and impact assessments, track mitigations and incidents, and map to regulatory frameworks with curated risk taxonomies. Framework mappings are vendor-curated interpretations, not regulator-endorsed.", "url": "https://trustible.ai/platform-overview/", "question": "Show me the intake-to-approval path for a high-risk use case, including which curated framework mappings and audit evidence it produces.", "reviewed": "2026-09-06" }, { "name": "TÜV SÜD ISO/IEC 42001 certification", "category": "Advisory & assurance services", "description": "TÜV SÜD audits and certifies AI management systems against ISO/IEC 42001, including risk-focused assessments of bias, privacy and security controls and optional integrated audits combining several management-system standards. The pages reviewed name no accreditation body for the AI scheme, so accreditation status could not be confirmed.", "url": "https://www.tuvsud.com/en/services/management-system-certification/iso-iec-42001", "question": "Demonstrate how an integrated ISO/IEC 42001 and ISO/IEC 27001 audit is planned for one AI platform, and what nonconformities you have typically raised on AI risk assessment.", "reviewed": "2026-09-06" }, { "name": "UiPath Maestro", "category": "Agent building & orchestration", "description": "UiPath's cloud orchestration layer that coordinates AI agents, RPA robots and people in BPMN, Flow or case-management models, with DMN decision tables. Governance is process-centric: role-based access, versioning and auditing of process and decision changes, rather than per-tool-call approval policy for agents.", "url": "https://docs.uipath.com/maestro/automation-cloud/latest/user-guide/overview", "question": "Demonstrate a Maestro BPMN process where an agent step hands an exception to a person, and show the versioned, audited history of the decision table change.", "reviewed": "2026-09-06" }, { "name": "Valence SaaS and AI Discovery", "category": "Agent discovery & inventory", "description": "Part of Valence's SaaS security platform: it inventories sanctioned and unsanctioned SaaS and AI applications, and continuously identifies OAuth tokens, API keys, connected apps and service accounts linking business SaaS tenants to third-party AI tools. Detection is API-based against connected SaaS tenants, so unconnected apps stay invisible.", "url": "https://www.valencesecurity.com/platform/saas-and-ai-discovery", "question": "Demonstrate how you surface an OAuth grant that connects an employee's unsanctioned AI tool to our Google Workspace tenant, including the scopes granted and the granting identity.", "reviewed": "2026-09-06" }, { "name": "ValidMind", "category": "Model lifecycle & governance", "description": "Model risk management and AI governance platform pairing a Python library that runs tests and generates model documentation with a review platform for validators, covering statistical, ML, LLM and agentic records with inventory, versioning and approval workflows aimed at regulated financial institutions.", "url": "https://validmind.com/platform/", "question": "Demonstrate the developer-to-validator handoff: library-generated documentation and test results entering a validation workflow with approvals and version control.", "reviewed": "2026-09-06" }, { "name": "Velatir", "category": "Work discovery & human delegation", "description": "European platform that inserts human approval into AI agent workflows: agents submit an operation via API, SDK or an MCP server integration, policies assess it, and low-risk requests auto-approve while higher-risk ones route to named human reviewers who approve, reject or request changes, with an audit trail. Reviewer coverage depends on integration work.", "url": "https://www.velatir.com/", "question": "Show me an agent tool call blocked at the MCP approval gate, the policy that routed it to a human, and the audit record of the reviewer's decision and reason.", "reviewed": "2026-09-06" }, { "name": "Verint Agent Factory", "category": "Work discovery & human delegation", "description": "Orchestration environment in Verint's CX platform for building and running a mixed workforce of human and AI agents: prebuilt and custom AI agents are connected to customer-experience workflows with explicit routing to humans when judgment is needed, plus centralized prompt and model governance. Scope is contact-centre work, not general enterprise task delegation.", "url": "https://www.verint.com/agent-factory/", "question": "Demonstrate a CX workflow where an AI agent hands off to a human agent, and show how the handoff condition and the governing prompt policy are configured.", "reviewed": "2026-09-06" }, { "name": "Veza AI Agent Security", "category": "Agent identity & access", "description": "Veza's access graph labels non-human accounts across its integrations, shows their effective permissions, supports right-sizing and periodic review, and its agent-focused release maps unmanaged AI agents and service accounts to human owners and enumerates the tools an agent may invoke. Agent capabilities were described as early access.", "url": "https://veza.com/use-cases/nhi-protection-and-governance/", "question": "Show me every tool and action an AI agent is authorized to invoke in a connected app, with the suggested human owner and a review campaign to revoke unused entitlements.", "reviewed": "2026-09-06" }, { "name": "W&B Registry", "category": "Model lifecycle & governance", "description": "Weights & Biases registry that curates versioned artifacts — models and datasets — into organisation-level collections with aliases, lineage tracking and access control, giving ML teams a single record of what is in production. Governance here means access and provenance, not risk or regulatory workflow.", "url": "https://docs.wandb.ai/models/registry", "question": "Show me how a model artifact is linked into a production collection, who can access it, and how lineage lets us reproduce that exact version.", "reviewed": "2026-09-06" }, { "name": "W&B Weave", "category": "Observability & traceability", "description": "Weights & Biases product for tracking LLM calls and application logic with automatic tracing and cost tracking, scorer-based evaluation and comparison tools, plus pre- and post-response safeguards. Platform controls include role-based access at team or project level, SSO via OIDC, SCIM provisioning and scoped service accounts.", "url": "https://weave-docs.wandb.ai/", "question": "Show me a traced LLM application with cost tracking, and demonstrate restricting project access to a named team using SSO-provisioned users.", "reviewed": "2026-09-06" }, { "name": "Workday Agent System of Record", "category": "Agent discovery & inventory", "description": "Workday functional area for cataloguing and administering an organisation's AI agents - Workday's own and third-party - with dedicated security domains for agent management, agent compliance and agent reporting. It is a registry and administration layer configured per tenant; evidence reviewed does not show detection of agents outside what is registered.", "url": "https://doc.workday.com/admin-guide/en-us/workday-ai/agents/set-up-agent-system-of-record.html", "question": "Demonstrate registering a third-party agent in ASOR and show the agent reporting and agent compliance views an auditor would rely on.", "reviewed": "2026-09-06" }, { "name": "Yields Model Risk Management", "category": "Model lifecycle & governance", "description": "Belgian model risk management software providing a configurable inventory of models, AI systems, agents, use cases, vendors and data sources with risk tiers, a workflow engine for validation and change control, automated validation reporting and monitoring, targeted at financial institutions with formal MRM functions.", "url": "https://www.yields.io/solutions/model-risk-management-software", "question": "Show me an automated validation report generated from the inventory record, including versioning history and the audit trail for a model change.", "reviewed": "2026-09-06" }, { "name": "Zenity AI Detection and Response (AIDR)", "category": "Agent security & threat detection", "description": "Runtime security layer for AI agents that analyses full execution sequences, including chained tool calls, retrievals and agent-to-agent handoffs, to detect direct and indirect prompt injection, unauthorised tool invocations and sensitive data leaving through agent activity, with inline blocking. Evidence is drawn from vendor platform pages.", "url": "https://zenity.io/platform/ai-detection-and-response", "question": "Show me AIDR detecting a slow-building indirect injection across several turns, blocking the unauthorised tool call it triggers, and the agent-to-agent handoff record.", "reviewed": "2026-09-06" } ]