source-of-truth
NewGive your coding agent a persistent memory. source-of-truth keeps a living docs/ catalog of what exists, why, and what rules apply — so AI stops rebuilding, deleting, and breaking things it never knew were there. Works with Claude, Codex, Cursor, Gemini, Copilot & more.
Summary
source-of-truth gives your coding agent a persistent memory by maintaining a living docs/ catalog of what exists, why, and what rules apply.
- It prevents AI from rebuilding, deleting, or breaking things it never knew were there, working seamlessly with Claude, Codex, Cursor, Gemini, Copilot, and more.
Install & Usage
mkdir -p .claude/skillsAdd the configuration to .claude/skills/source-of-truth.md
/source-of-truthUse Cases
Usage Examples
/source-of-truth update docs/architecture.md with the new microservice endpoints
Add a rule to docs/rules.md that says 'Never delete files in the config/ directory'
Read docs/decisions.md to understand why we chose PostgreSQL over MongoDB
Security Audits
Frequently Asked Questions
What is source-of-truth?
source-of-truth gives your coding agent a persistent memory by maintaining a living docs/ catalog of what exists, why, and what rules apply. It prevents AI from rebuilding, deleting, or breaking things it never knew were there, working seamlessly with Claude, Codex, Cursor, Gemini, Copilot, and more.
How to install source-of-truth?
To install source-of-truth: create the skills directory (mkdir -p .claude/skills), then add the config to .claude/skills/source-of-truth.md. Finally, /source-of-truth in Claude Code.
What is source-of-truth best for?
source-of-truth is a other categorized under Documentation. It is designed for: agent. Created by ngocquang.
What can I use source-of-truth for?
source-of-truth is useful for: Documenting project architecture and design decisions so AI agents don't accidentally overwrite critical files.; Maintaining a shared rulebook for AI agents to follow consistent coding conventions across a team.; Tracking why certain workarounds or technical debt exists to prevent AI from 'fixing' them incorrectly.; Recording API contracts and data flow diagrams so agents can generate accurate integration code.; Logging deployment configurations and environment variables to avoid breaking production setups.; Creating a changelog of agent actions to audit and revert changes if needed..