
Microsoft toolkit for policy enforcement, identity and audit of autonomous agents
- Open source
- Open-source agent
- 6.6K GitHub stars
- Security

Microsoft toolkit for policy enforcement, identity and audit of autonomous agents

Open-source gateway adding identity, policy and audit between AI agents and your systems
Experimental agent wallet that seals agreements in a public register anchored in Bitcoin
Runs locallyPassive audit that checks whether an MCP endpoint exposes its tools without authentication
Runs locallySign agent actions into a tamper-evident log and verify any agent's log from an MCP client

Ask why Korzen Bot Protection blocked a request and which AIs read your site

Read-only checks of an AI agent's Sigvara standing, score, stake and slashes
Runs locallyA safer MCP for Dokploy: agents deploy and fix apps without seeing keys or secrets
Not checked yetScan PDF, Office, RTF and HTML files for hidden text before an agent reads them
MCP wrapper that holds database writes until a second person approves them on their phone
Runs locally
Check children's screen time and manage device rules from an AI assistant
Not checked yetQuery software inventory, vulnerabilities and end-of-life components

Open-source control plane that firewalls MCP tool calls and caps agent model spend

Read-only payment preflight for AI agents: pay, review or do-not-pay before signing
Check an AI agent's action against signed rules before it commits, with auditable proof
Run an 88-test MCP spec compliance suite and get an A-F grade from CLI or agent
Runs locallyLocal input firewall that scans text, files and MCP metadata for prompt injection
Runs locallySealed microVM sandboxes for agents to run untrusted code, locally or in the cloud
Not checked yetnpm registry intelligence, security audits and publish-side writes without the 2FA prompt
Runs locallyScreen an x402 payment counterparty with allow, review or block before an agent pays
Runs locallyRoute agent LLM calls through spending caps, metering and verifiable receipts
Not checked yet
Hash and sign evidence of what AI agents did, mapped to AI governance frameworks
Runs locallyFree read-only server listing known security incidents and fix procedures for dev software
Not checked yet
Read-only MCP gateway that approves exact tools and blocks changed or unknown calls
Not checked yetModerate game and community chat from your agent, with a no-signup test key
Not checked yetStatic security scanner for MCP servers, agent skills and npm or PyPI packages
Runs locallyYouTube video analysis, C2PA image marking and metadata removal, paid per call
Not checked yetPublic facts on a French phishing-awareness training service: plans, templates, security
Not checked yetLook up safety scores, incidents and verified results for AI trading bots and tools
Not checked yet