Skip to content

feat: add 5 engineering-adjacent skills (conflicts, cso, risk, escalation, ai-hybrid)#106

Open
HMAKT99 wants to merge 2 commits intogarrytan:mainfrom
HMAKT99:arun/engineering-skills
Open

feat: add 5 engineering-adjacent skills (conflicts, cso, risk, escalation, ai-hybrid)#106
HMAKT99 wants to merge 2 commits intogarrytan:mainfrom
HMAKT99:arun/engineering-skills

Conversation

@HMAKT99
Copy link

@HMAKT99 HMAKT99 commented Mar 16, 2026

Summary

  • Adds 5 new skills focused on engineering operations: conflict detection, security auditing, risk management, incident response, and AI-human collaboration
  • Each skill follows a distinct persona with structured analysis frameworks
  • Fully integrated with the existing template pipeline, test suite, and CI freshness checks

New Skills

Skill Persona Purpose
/conflicts Tech Lead Cross-PR semantic conflict prediction, merge ordering
/cso Chief Security Officer OWASP Top 10 + STRIDE threat modeling
/risk Chief Risk Officer Risk register with likelihood × impact scoring
/escalation Escalation Manager Incident triage, war rooms, post-incident reviews
/ai-hybrid AI Architect Human-AI task splitting, workflow optimization

Test plan

  • `bun run gen:skill-docs --dry-run` — all skills FRESH
  • All existing tests pass with new skills registered
  • Each skill has valid YAML frontmatter, preamble, and structured outputs
  • All skills are read-only (no Edit in allowed-tools)

🤖 Generated with Claude Code

…tion, ai-hybrid)

New skills for engineering workflows:
- /conflicts — Cross-PR semantic conflict predictor (Tech Lead persona)
- /cso — OWASP Top 10 + STRIDE threat modeling (Chief Security Officer)
- /risk — Risk register with likelihood x impact scoring (Chief Risk Officer)
- /escalation — Incident triage, war rooms, post-incident reviews (Escalation Manager)
- /ai-hybrid — Human-AI task splitting and workflow optimization (AI Architect)
@HMAKT99 HMAKT99 force-pushed the arun/engineering-skills branch from 36bc658 to ce9b313 Compare March 16, 2026 16:55
Tier 3 evals (~$0.10/run) using Claude Sonnet as judge:
- conflicts: conflict detection workflow quality
- cso: security audit methodology quality
- risk: risk assessment methodology quality
- escalation: incident response workflow quality
- ai-hybrid: AI collaboration workflow quality
- Cross-skill consistency check (read-only patterns)

Run: EVALS=1 bun test test/new-skills-llm-eval.test.ts
Requires: ANTHROPIC_API_KEY
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant