The Security Kernel for AI Agents — MCP/A2A gateway with policy enforcement, taint tracking, sandboxed execution, deterministic envelopes, and Sigstore audit. OWASP ASI 2026 compliant.
Latest release v0.3.0 — v0.3.0 — Deep-Advance Hardening · 25 May 2026
8 files for GitHub Copilot, Claude Code and Codex: mcpkernel copilot-instructions.md, mcpkernel policy-writing.instructions.md, mcpkernel python-conventions.instructions.md, mcpkernel testing-guidelines.instructions.md and 4 more — 2,161 tokens loaded in every session.
.github/copilot-instructions.md A 1,344 tok .github/instructions/policy-writing.instructions.md A 220 tok .github/instructions/python-conventions.instructions.md A 267 tok .github/instructions/testing-guidelines.instructions.md A 330 tok .github/skills/docs-sync/SKILL.md A 47 tok .github/skills/research-and-improve/SKILL.md A 48 tok .github/skills/search-and-fix/SKILL.md A 45 tok .github/skills/test-and-merge/SKILL.md A 50 tok These files are piyushptiwari1/mcpkernel's own configuration — they tell GitHub Copilot, Claude Code and Codex how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.