defense-in-depth / skills
AGENTS.md4 starsChanged 7 months agoMITView on GitHub
Read this when: You are an AI agent contributing to this project and need to load specialized behavioral guidelines for your task. Skip if: You are a human developer — skills are behavioral guidelines for AI contributors, not developer documentation. See docs/ instead. Skills are lazy-loaded behavioral specifications for AI agents. They define how an agent should approach a specific type of task, what constraints apply, and what outputs are required. Skills are never imported by TypeScript source code. They…
In this repository16 other files it gives its agents.
- .agents/AGENTS.mdAGENTS.md
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- skill-ai-dspy-validatorSkill
- Bootstrap AgentSkill
- CreatorSkill
- Deep ResearchSkill
- skill-devops-github-actionsSkill
- skill-guard-governanceSkill
- skill-impact-predictorSkill
- skill-review-codeSkill
- Self ReflectionSkill
- skill-surgical-refactorerSkill
- skill-test-architectSkill
- skill-threat-modeling-expertSkill
- [Skill Name]Skill
The file
Files
.agents/skills — Agent Skills for defense-in-depth
Read this when: You are an AI agent contributing to this project and need to load specialized behavioral guidelines for your task.
Skip if: You are a human developer — skills are behavioral guidelines for AI contributors, not developer documentation. See
docs/instead.
What are Skills?
Skills are lazy-loaded behavioral specifications for AI agents. They define how an agent should approach a specific type of task, what constraints apply, and what outputs are required.
Skills are never imported by TypeScript source code. They are read by AI agents at task time — not by the runtime engine.
Why not AAOS captain skills? Captain skills (epistemologist, socratic, planner...) are deeply coupled to the AAOS multi-agent platform and its orchestration vocabulary. They are not portable to Devin, Jules, CodeRabbit, or other external agents that work in this repo. DiD ships specialist skills only — domain-specific, self-contained, and usable by any AI agent without AAOS context.
Skill Discovery
Load only the skill relevant to your current task. Do not load all skills.
Your task type → Load this skill
─────────────────────────────────────────────────────────────────────────
Writing or modifying a Guard → skill-guard-governance/SKILL.md
Threat modeling a new check → skill-threat-modeling-expert/SKILL.md
Writing adversarial tests → skill-test-architect/SKILL.md
Code review of a PR → skill-review-code/SKILL.md
Refactoring existing guard code → skill-surgical-refactorer/SKILL.md
Mapping blast radius of a change → skill-impact-predictor/SKILL.md
CI/CD pipeline or GitHub Actions → skill-devops-github-actions/SKILL.md
DSPy evaluation quality / tuning → skill-ai-dspy-validator/SKILL.mdSkill Index
| Skill | Domain | When to load |
| skill-guard-governance | Governance | Writing / registering a Guard |
| skill-threat-modeling-expert | Security | Designing threat coverage |
| skill-test-architect | Testing | Adversarial test design |
| skill-review-code | Quality | PR code review |
| skill-surgical-refactorer | Refactoring | Safe guard refactors |
| skill-impact-predictor | Governance | Pre-implementation blast-radius analysis |
| skill-devops-github-actions | CI/CD | GitHub Actions / CI gates |
| skill-ai-dspy-validator | Intelligence | DSPy eval quality |
Skill Standards
See SKILL_STANDARDS.md for the contract every skill must satisfy.
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.

