Mneme is open-source architectural drift prevention for the agentic AI SDLC. It turns architectural decisions, ADRs, and engineering standards into deterministic guardrails that keep AI coding agents aligned with repository architecture. Architectural governance is the mechanism: approved decisions become enforceable before incompatible changes reach review, with verdicts linked to their governing source. Native integrations: Claude Code, Claude Agent SDK, Google Antigravity, and Codex CLI. Validated compatibility: Paperclip. Rules export: Cursor. CLI-based CI gates: GitHub Actions and GitLab CI. Experimental: OpenCode and Kiro. Install: pip install mneme-hq. Docs: https://mnemehq.com/docs/ Integration matrix: https://mnemehq.com/integrations/ Source: https://github.com/MnemeHQ/mneme
Features
- Native integrations: Claude Code, Agent SDK, Antigravity, and Codex CLI
- Validated compatibility: Paperclip
- Rules export: Cursor; CI gates: GitHub Actions and GitLab CI
- Experimental: OpenCode and Kiro