Inkog
Scans AI agent code for security vulnerabilities and logic flaws as you build, right inside your Claude or Cursor conversation. Exposes tools for static analysis (inkog_scan), AGENTS.md governance verification, MCP server auditing, and EU AI Act compliance reporting. The verify_governance tool is the standout here: it checks whether your governance declarations actually match what the code does. You can ask Claude to scan a file, explain findings in plain English, then apply fixes without leaving the chat. Also audits agent-to-agent delegation in multi-agent systems for infinite loops and privilege escalation. Works with LangChain, CrewAI, AutoGen, and custom implementations. Requires an API key from app.inkog.io.
- GitHub stars
- 2
- tools
- 7
- auth
- Secret needed
- transport
- STDIO
- env var
- 1
- last updated
- Feb 28, 2026
Tools
7 toolsPublic tool metadata for what this MCP can expose to an agent.
inkog_scanStatic analysis for AI agent code - finds logic flaws and security risks.
Static analysis for AI agent code - finds logic flaws and security risks.
No parameter schema in public metadata yet.
inkog_verify_governanceValidate that AGENTS.md declarations match actual code behavior.
Validate that AGENTS.md declarations match actual code behavior.
No parameter schema in public metadata yet.
inkog_compliance_reportGenerate compliance reports for regulatory frameworks.
Generate compliance reports for regulatory frameworks.
No parameter schema in public metadata yet.
inkog_explain_findingGet detailed explanation and remediation guidance for a security finding.
Get detailed explanation and remediation guidance for a security finding.
No parameter schema in public metadata yet.
inkog_audit_mcp_serverSecurity audit any MCP server from the registry or GitHub.
Security audit any MCP server from the registry or GitHub.
No parameter schema in public metadata yet.
inkog_generate_mlbomGenerate a Machine Learning Bill of Materials listing all AI components.
Generate a Machine Learning Bill of Materials listing all AI components.
No parameter schema in public metadata yet.
inkog_audit_a2aAudit Agent-to-Agent communications for security risks.
Audit Agent-to-Agent communications for security risks.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
INKOG_API_KEY*secretYour Inkog API key from https://app.inkog.io (free tier available)
Readme
Inkog MCP Server
Security companion for AI agent development in Claude, Cursor, and Claude Code.
Ask your AI pair-programmer to build an agent. Inkog checks it as you code — scanning for vulnerabilities, explaining findings in plain English, verifying AGENTS.md governance, and auditing agent-to-agent delegation. All inside the same conversation, no context switch.
Available in Claude Desktop, Cursor, Claude Code, ChatGPT, and any MCP-compatible client.
The Dev-Flow Loop
Inkog is designed to live inside the conversation where you build the agent — not as a post-hoc gate:
Details
- Repository
- github.com/inkog-io/inkog-mcp
- Category
- Security & Pentesting
- Package
@inkog-io/mcp- Transport
- STDIO
- Auth
- Required
- Registry
- MCP registry, active
Related Security & Pentesting MCP servers
View all
LLM Assurek11-software-solutions/k11techlab-llm-assurance-toolkitRed-team LLM apps: 41 OWASP LLM Top 10 probes, adaptive attacks, CI gate, audit reports.0stars
Webcheckyifanyifan897645/webcheck-mcpWebsite health analysis: SEO, accessibility, performance, security, and broken links2stars
Lintsurelintsure/lintsureCompliance readiness checks for code: EU AI Act, GDPR, OWASP, CWE, Cyber Resilience Act. Local.2stars
MCPAmpel - MCP Security Scannermcpampel/mcpampelScan installed MCP servers for security vulnerabilities with 16 detection engines.2stars
HoneyLabshoneylabshq/honeylabs-mcpHoneypot probe data: IP reputation, scanners, CVE probing, TLS and SSH fingerprints.2stars
Compliance Trestle Mcposcal-compass/compliance-trestle-mcpAn MCP server that provides tools to author OSCAL security compliance documentation2stars