spec-validator
Skill0 starsChanged 4 months agoMITView on GitHub
Validates engineering specs before implementation. Ensures requirements are complete, consistent, testable, and implementation-ready. Use for tickets, PRDs, RFCs, or verbal feature asks. Emits SPEC_STATUS. Never redesigns the product or starts coding.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- fernandopetry/keelsontheir own version, first seen Oct 23, 2026
- EmeaAppGbb/spec2cloudtheir own version, first seen Oct 9, 2026
- crgarcia12/Liliputtheir own version, first seen Jul 14, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- EmeaAppGbb/spec2cloud★ 100same name
- fernandopetry/keelson★ 5same name
- crgarcia12/Liliput★ 4same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository27 other files it gives its agents.
- adr-enforcerSkill
- ai-eval-harnessSkill
- api-contract-guardianSkill
- change-impact-analyzerSkill
- code-reviewerSkill
- context-loaderSkill
- defect-analystSkill
- delivery-plannerSkill
- engineering-mentorSkill
- engineering-osSkill
- feature-implementerSkill
- guardrails-configuratorSkill
- incident-commanderSkill
- loop-engineerSkill
- migration-surgeonSkill
- model-routerSkill
- observability-designerSkill
- performance-auditorSkill
- postmortem-writerSkill
- pr-generatorSkill
- quality-gateSkill
- refactoring-advisorSkill
- requirements-grillSkill
- security-auditorSkill
- tdd-executorSkill
- test-strategy-designerSkill
- threat-modelerSkill
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 45
Do not invent product requirements.
- Distinctive: line 46
Prefer Block over hopeful Ready.
- Distinctive: line 112
Never “fill in” product decisions without labeling them as assumptions needing approval.
- Distinctive: line 113
Never mark Ready when ACs are not testable.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Purpose
Determine whether work is safe to plan and implement. Ambiguity is a defect.
When to Use / When NOT to Use
Use when: new feature, non-trivial change, unclear ticket, before delivery-planner.
Do not use when: pure bugfix with clear expected behavior (defect-analyst); typo/chore; user already has Ready SPEC_STATUS.
Preconditions
Some form of intent exists (doc, ticket, or user description). If absent, interview briefly then validate.
Inputs / Outputs
Inputs: spec/ticket/ask, optional CONTEXT_PACK.
Outputs: SPEC_STATUS with AC IDs.
Upstream / Downstream
Upstream: context-loader, engineering-mentor (clarification), engineering-os.
Downstream: adr-enforcer, delivery-planner, test-strategy-designer.
Core Principles
- Ambiguity is a defect.
- Acceptance criteria must be testable.
- Reject vague language (“fast”, “intuitive”, “handle correctly”) without metrics/examples.
- Tier validation effort to risk.
- Security/privacy NFRs required for auth/PII/payments/integrations.
- Do not invent product requirements.
- Prefer Block over hopeful Ready.
Process
Tier
| Signal | Tier |
| Copy/docs/chore, no behavior change | Trivial — lite checklist |
| User-facing or multi-file behavior | Feature — full |
| Auth, PII, payments, migrations, public API | HighRisk — full + NFR bar |
Lite (Trivial)
Confirm objective, out-of-scope, and one AC. Ready if unambiguous.
Full
- Completeness — actors, flows, data, errors, edge cases, out of scope.
- Consistency — no contradictions.
- Testability — each AC observable/verifiable; assign
AC-#IDs. - Constraints — perf, compatibility, deadlines.
- NFRs — for HighRisk: authz, data handling, audit, failure modes.
- Open questions — blocking vs non-blocking.
- Readiness decision.
Evidence Requirements
Quote ambiguous phrases. Map each blocking issue to a missing AC or contradiction.
Stop Conditions / Failure Modes
| Condition | Action |
| Blocking ambiguity remains | NotReady / Blocked |
| HighRisk missing security NFR section | NotReady |
| Spec asks for architecture change | Flag ADR need; do not approve silently |
Severity + Confidence
Severity on issues (Critical = unsafe to build). Confidence on readiness judgment.
Output Contract
## SPEC_STATUS
Tier: Trivial | Feature | HighRisk
Readiness: Ready | ReadyWithConditions | NotReady | Blocked
Acceptance criteria:
- AC-1: ...
Blocking issues: ...
Non-blocking issues: ...
NFR notes: ...
Decision: Proceed | ProceedWithConditions | Revise | BlockHandoffs
- delivery-planner — when Ready / ReadyWithConditions
- adr-enforcer — architecture implications
- requirements-grill (if installed) — adversarial clarification
- security-auditor — threat-sensitive specs (advisory)
Never
- Never start implementation.
- Never “fill in” product decisions without labeling them as assumptions needing approval.
- Never mark Ready when ACs are not testable.
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.