prd-implementation-precheck
Skill85 starsChanged 6 months agoApache-2.0View on GitHub
Use this skill at Step 1 (PRD pre-check) and Step 2 (TDD alignment) of the v2 12-step SOP to find blockers and warnings in PRD / TDD before any implementation work begins. Produces a structured gap list pointing to missing scope boundaries, undefined acceptance criteria, unresolved cross-team dependencies, vague metrics, and stage breakdown gaps. Chinese trigger examples: "PRD 预检", "PRD 检查", "TDD 对齐检查", "需求文档审查", "PRD 缺口扫描", "Step 1", "Step 2". Do NOT use to write the PRD itself (use `prd` Skill), do NOT use as substitute for compliance precheck (Step 0 is separate). Success = zero blockers, zero warnings, stage breakdown present, every acceptance criterion has a measurable test hook, every dependency has a named owner.
In this repository24 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- ask-questions-if-underspecifiedSkill
- audit-snapshot-microSkill
- audit-snapshotSkill
- compliance-precheckSkill
- compliance-reviewSkill
- memory-curateSkill
- mocking-stubbingSkill
- openspec-apply-changeSkill
- openspec-continue-changeSkill
- openspec-new-changeSkill
- openspec-verify-changeSkill
- phi-desensitizeSkill
- prd-authorSkill
- prd-precheckSkill
- prdSkill
- prompt-injection-scanSkill
- quick-fixSkill
- requesting-code-reviewSkill
- systematic-debuggingSkill
- task-decompositionSkill
- tdd-alignmentSkill
- test-data-generationSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 33
Every feature has a defined boundary (in/out of scope, both stated)
- Distinctive: line 34
Every acceptance criterion has a measurable test hook
- Distinctive: line 35
Every cross-team dependency has a named owner
- Distinctive: line 36
Every metric has a baseline + target + measurement method
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill 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
PRD Implementation Precheck
The Step 1 / Step 2 gate. Catches PRD / TDD ambiguity before OpenSpec is generated, so we don't have to retrofit later.
Core mental model
A PRD is implementation-ready when:
- Every feature has a defined boundary (in/out of scope, both stated)
- Every acceptance criterion has a measurable test hook
- Every cross-team dependency has a named owner
- Every metric has a baseline + target + measurement method
- The scope is broken into shippable stages, each with its own DoD
This skill scans for violations and emits a gap report.
What this skill produces
PRD_PRECHECK_REPORT.md with:
- Blockers (must fix before proceeding)
- Warnings (should fix; can proceed with explicit override)
- Suggestions (informational)
When NOT to use
- Writing or expanding a PRD (use
prd) - Compliance pre-check (use
compliance-precheckStep 0) - Architecture review (use Plan agent / openspec-new-change)
Active context bundle
Always load first
- This
SKILL.md - The target PRD / TDD file
reference/precheck-checklist.md— 30+ canonical gapsreference/severity-rubric.md
Workflow
- Read PRD / TDD top to bottom.
- Run checklist — for each item, mark present / missing / vague.
- Stage breakdown audit — does the PRD propose a stage 1 / 2 / 3 split? Each stage has its own DoD?
- Metric audit — every KPI has baseline + target + measurement method?
- Dependency audit — every "needs X from team Y" has an owner + due date?
- Compose report — group by severity, point at exact PRD section/line.
- Hand off — if blockers > 0, send back to
prdSkill; if only warnings, surface to user for explicit accept/fix decision.
Hard gate / soft gate
Hard for blockers. Warnings can be accepted with sign-off (recorded in report).
Common failure modes
- "Looks complete" trap — PRD reads fluently but lacks measurable criteria. Mitigation: always demand metric formula, not metric name.
- Stage breakdown handwaved — "Stage 1 MVP, Stage 2 polish" with no scope cut. Mitigation: require feature-level allocation per stage.
- Owner = team name — "Backend team" is not an owner. Mitigation: require named individual + role.
- Acceptance via dogfooding — "users say it's good" is not measurable. Mitigation: require quantitative + qualitative pair.
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.

