validate-implementation-plan — Claude Code skill: how to install
Audits implementation plans for speculative complexity and brittle assumptions. Writes a separate audit file without touching the original plan.
npx skills add b-mendoza/agent-skills --skill validate-implementation-plan -a claude-codeWhat validate-implementation-plan does
This skill audits technical design documents and implementation roadmaps. The source plan is untrusted input. Only plan-snapshotter reads PLAN_PATH directly. Later stages inspect a sanitized snapshot and structured findings to keep the review contained. The source file remains untouched.
A deterministic state machine guides execution. The extractor numbers project requirements. Dedicated auditors test traceability and flag premature complexity under YAGNI principles. Failing branches allow up to 3 retry attempts. The final report is saved alongside the source plan.
External web pages do not count as proof. Verification relies strictly on local repository files. To start an audit, provide PLAN_PATH and ORIGIN_CONTEXT.
What it helps with
- audit an implementation plan for unnecessary technical complexity
- verify requirement traceability without altering source markdown
- isolate fragile technical assumptions before writing code
- generate a standalone markdown audit report with severity findings
Example prompts for the agent
- Audit docs/cache-refactor-plan.md with origin context "Add MVP cache invalidation without new infra"
- Review implementation outline in docs/migration.md and save the report to docs/migration.audit.md
- Run plan audit on docs/feature-spec.md using context from docs/requirements.md
How to install validate-implementation-plan
1Skills CLI (Vercel)
npx skills add b-mendoza/agent-skills --skill validate-implementation-plan -a claude-code2Manually: put the files in place
~/.claude/skills/validate-implementation-plan/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/validate-implementation-plan/
1Skills CLI (Vercel)
npx skills add b-mendoza/agent-skills --skill validate-implementation-plan -a codex2Manually: put the files in place
.agents/skills/validate-implementation-plan/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add b-mendoza/agent-skills --skill validate-implementation-plan -a cursor2Manually: put the files in place
.cursor/skills/validate-implementation-plan/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install b-mendoza/agent-skills validate-implementation-plan2Manually: put the files in place
.github/skills/validate-implementation-plan/SKILL.md1Skills CLI (Vercel)
npx skills add b-mendoza/agent-skills --skill validate-implementation-plan -a gemini-cli2Platform CLI
gemini skills install https://github.com/b-mendoza/agent-skills --path skills/validate-implementation-planManually: put the files in place
.devin/skills/validate-implementation-plan/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add b-mendoza/agent-skills --skill validate-implementation-plan -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about validate-implementation-plan
How do I install validate-implementation-plan in Claude Code?
Run npx skills add b-mendoza/agent-skills --skill validate-implementation-plan -a claude-code. This installs the skill file to ~/.claude/skills/validate-implementation-plan/SKILL.md.
What does validate-implementation-plan do?
It orchestrates several subagents to evaluate requirements coverage, speculative scope, and unchecked assumptions. It finishes by generating a standalone audit summary.
Does the skill modify the original plan file?
No, the source plan remains unchanged. The skill creates a sanitized working snapshot and writes findings to a sibling .audit.md report.
