Skip to content

Korext: AI Code Governance

by Korext · remote · local (npm) · no sign-in to list tools · v3.0.1 · published by GitHub user Korext

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

Changes · 2
CompareRegistry entry

5 tools: 2 change things and 3 only read.

Scanned Oct 21, 2026

No check matched anything in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http governance https://app.korext.com/api/mcp

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

5 tools

Changes things 2Claude Code: ask
  • check_codeCheck a code snippet against one or more policy packs.No findings

    Why it is in this group

    • The input takes a statement or command to run (input.code), which can change anything it can reach.
    Description, as the server wrote it
    Check a code snippet against one or more policy packs. Returns violations with severity, governance context (CWE, MITRE, regulatory), and line numbers.
  • generate_proofScan code against policy packs and generate a cryptographically signed proof bundle.No findings

    Why it is in this group

    • The name "generate_proof" says "generate": a word for a change.
    • The input takes a statement or command to run (input.code), which can change anything it can reach.
    Description, as the server wrote it
    Scan code against policy packs and generate a cryptographically signed proof bundle. Returns the bundle ID, decision (PASS/BLOCK), violation count, and verification URL.
Only reads 3Claude Code: allow
  • get_directivesGet enriched policy directives and living standards for one or more policy packs.No findings

    Why it is in this group

    • The name "get_directives" says "get": a word for reading.
    Description, as the server wrote it
    Get enriched policy directives and living standards for one or more policy packs. Returns rules with governance context (CWE, MITRE, regulatory references) for inference time enforcement.
  • get_proofRetrieve an existing proof bundle by ID.No findings

    Why it is in this group

    • The name "get_proof" says "get": a word for reading.
    Description, as the server wrote it
    Retrieve an existing proof bundle by ID. Returns the decision, violations, packs, and verification status.
  • list_packsList available policy packs with industry and region tags.No findings

    Why it is in this group

    • The name "list_packs" says "list": a word for reading.
    Description, as the server wrote it
    List available policy packs with industry and region tags.

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Install

Pinned to version 3.0.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http governance https://app.korext.com/api/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__governance__get_directives",
      "mcp__governance__list_packs",
      "mcp__governance__get_proof"
    ],
    "ask": [
      "mcp__governance__check_code",
      "mcp__governance__generate_proof"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. It assumes the server is installed as governance. Review it before you use it.