codeaudit
llms.txt1 starsChanged 6 months agoMITView on GitHub
Use this file to discover the available CodeAudit documentation before exploring further. - README: ../README.md Overview, setup, safety model, skills pack, and example workflow. - Agent operating manual: ../AGENTS.md Required read order, engineering rules, security rules, workflow, and Git rules. - Tool policy: ../TOOLS.md Allowed MCP tools, disallowed actions, capability boundaries, and future tool rules. - Agent soul: ../SOUL.md CodeAudit tone, priorities, and operating boundaries. - MCP clients: clients.md Local stdio and Streamable HTTP setup examples for OpenAI Codex, Claude…
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- ai-app-security-reviewSkill
- codeaudit-orchestratorSkill
- codeauditSkill
- docs-claims-evidence-reviewSkill
- enterprise-code-qualitySkill
- github-issue-pr-workflowSkill
- nextjs-security-reviewSkill
- official-docs-groundingSkill
- python-backend-qualitySkill
- python-mcp-server-qualitySkill
- refactor-with-testsSkill
- skill-supply-chain-auditorSkill
The file
We split this llms.txt into the rules it gives an agent
8 rules. About 0 are common, 7 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 3
Use this file to discover the available CodeAudit documentation before exploring further.
- Distinctive: line 36
Read-only safety model, redaction, filesystem boundaries, and skill audit boundaries.
- Distinctive: line 67
Read README, AGENTS, TOOLS, and this index.
- Distinctive: line 69
Call detect_project on the target project.
- Distinctive: line 70
Call route_skills with the user task and detected project.
- Distinctive: line 71
Follow workflowPhases, recommendedToolSequence, skillActivationOrder, and qualityGates.
- Distinctive: line 72
Run relevant audit tools before edits in existing projects.
- Distinctive: line 73
Do not claim secure, production-ready, enterprise-grade, or fully tested without evidence.
≈ “Common” is an estimate: at least 1 in 5 of the 3,592 llms.txt files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
# CodeAudit MCP Documentation Index Use this file to discover the available CodeAudit documentation before exploring further. ## Start Here - README: ../README.md Overview, setup, safety model, skills pack, and example workflow. - Agent operating manual: ../AGENTS.md Required read order, engineering rules, security rules, workflow, and Git rules. - Tool policy: ../TOOLS.md Allowed MCP tools, disallowed actions, capability boundaries, and future tool rules. - Agent soul: ../SOUL.md CodeAudit tone, priorities, and operating boundaries. ## MCP Setup - MCP clients: clients.md Local stdio and Streamable HTTP setup examples for OpenAI Codex, Claude Code, Cursor, VS Code, Claude Desktop, Gemini CLI, Cline/Roo/Windsurf-style clients, and MCP Inspector. - MCP tool contracts: mcp-tool-contracts.md Tool inputs, outputs, output rules, and schema expectations. - Deployment: deployment.md Free public npm distribution, production HTTP deployment, Docker, environment variables, verification, and current limits. ## Design And Safety - Architecture: architecture.md Server structure, schemas, tools, policies, skills, and data flow. - Security model: security-model.md Read-only safety model, redaction, filesystem boundaries, and skill audit boundaries. - Threat model: threat-model.md Main threats, mitigations, and residual risks. - Skill routing design: skill-routing-design.md Route manifest fields, workflow phases, routing inputs, and workflow intent. - Official docs integration: official-docs-integration.md How CodeAudit routes agents to current official documentation. - Roadmap: roadmap.md Planned improvements and future capabilities. ## Built-In Skills - codeaudit: ../skills/codeaudit/SKILL.md - codeaudit-orchestrator: ../skills/codeaudit-orchestrator/SKILL.md - python-backend-quality: ../skills/python-backend-quality/SKILL.md - python-mcp-server-quality: ../skills/python-mcp-server-quality/SKILL.md - enterprise-code-quality: ../skills/enterprise-code-quality/SKILL.md - nextjs-security-review: ../skills/nextjs-security-review/SKILL.md - ai-app-security-review: ../skills/ai-app-security-review/SKILL.md - docs-claims-evidence-review: ../skills/docs-claims-evidence-review/SKILL.md - refactor-with-tests: ../skills/refactor-with-tests/SKILL.md - github-issue-pr-workflow: ../skills/github-issue-pr-workflow/SKILL.md - official-docs-grounding: ../skills/official-docs-grounding/SKILL.md - skill-supply-chain-auditor: ../skills/skill-supply-chain-auditor/SKILL.md ## Recommended Agent Workflow 1. Read README, AGENTS, TOOLS, and this index. 2. Configure the MCP server using docs/clients.md. 3. Call detect_project on the target project. 4. Call route_skills with the user task and detected project. 5. Follow workflowPhases, recommendedToolSequence, skillActivationOrder, and qualityGates. 6. Run relevant audit tools before edits in existing projects. 7. Do not claim secure, production-ready, enterprise-grade, or fully tested without evidence.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.