"GitHub Actions" matching MCP connectors:

Portable identity, authority, trust receipts, and Vault secrets before sensitive AI actions.

Fail-closed action authorization, MCP risk scanning, x402 checks, and signed receipts.

Policy review and purchase discovery for AI-agent commerce actions.

Post-quantum, tamper-evident receipts for agent actions. Ed25519 + ML-DSA-65, offline verify.

MCP enforcement layer that intercepts AI agent actions and blocks rule violations before execution.

Issue & verify signed (ed25519), hash-chained, timestamped provenance receipts for agent actions.

Governed AI actions with signed, verifiable receipts: free keyless reads, human-approved writes.

Governance copilot for AI-assisted coding. 72 packs, 532 rules, proof bundles.

Real engineering memory marketplace for autonomous AI agents. Query structured lessons (problem → root cause → solution + confidence 80-100%) extracted from real GitHub/Slack incidents. Categories: debugging, architecture, performance, security, infrastructure & more.

Chat Thing is the fastest way to build AI-native customer support. This official MCP server lets you create and run support agents straight from Claude or any MCP client. Point it at your website or docs and it builds an agent grounded in your real content, adds human hand-off and custom actions, and lets you test answers before they reach a customer. No dashboard required.

Free selector plus read-only GitHub bounty, agent harness, Actions, flake, and MCP drift decisions.

Read-only GitHub bounty, agent harness, Actions failure, flake, and MCP tool-drift decisions.

Bitcoin-anchored, tamper-evident audit log for AI agents — record, disclose and verify actions.

Hosted MCP server for noticed. Exposes ~50 noticed agent capabilities (network search across GitHub + LinkedIn, missions, PRM, memory, workspace, web search, cron, ...) via two meta-tools — search + execute. Bearer-API-key auth; key minted at noticed.so/dashboard/api-keys.

An MCP server that provides Javelin Standalone Guardrails