spec-to-code-verification
Skill1 starsChanged 7 days agoMITView on GitHub
Compare testable requirements with implementation and test evidence, producing a requirement-to-evidence matrix. Use for a deliberate specification audit, not for every ordinary code edit.
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- adaptive-workflowSkill
- architecture-mapSkill
- browser-diagnosticsSkill
- domain-modelingSkill
- feature-discoverySkill
- lean-senior-engineeringSkill
- release-readinessSkill
- research-plan-implementSkill
- skill-security-guardSkill
- systematic-debuggingSkill
- transaction-safe-engineeringSkill
- ui-design-intelligenceSkill
- ui-implementationSkill
- ui-quality-verificationSkill
- verification-before-completionSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 18
Do not invent missing requirements.
- Distinctive: line 30
Use code search and existing tests.
- Distinctive: line 35
Read traceability-matrix.md for the output format.
≈ “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
Specification-to-Code Verification
Determine whether the implementation actually satisfies the written requirements. This is a verification workflow; it produces findings unless implementation is also requested.
Establish the requirement set
Locate the authoritative specification, acceptance criteria, or contract. Record its version and scope. Split compound requirements into individually testable statements with stable IDs. Separate must-have requirements from examples, non-goals, and unresolved questions.
Do not invent missing requirements. If the spec is ambiguous or contradictory, report
the ambiguity and route product decisions to feature-discovery.
Trace to evidence
For each requirement, collect:
- implementation location or external contract;
- automated test or executable check;
- observed result and environment;
- gaps, partial coverage, and assumptions.
Use code search and existing tests; do not count a test name as proof without checking what it asserts. Distinguish unit, integration, contract, browser, and manual evidence. A requirement with no test can still be verified manually, but label the method and residual risk.
Read traceability-matrix.md for the output format.
Report and verify fixes
Report pass, fail, partial, untested, and out-of-scope requirements. Include orphan tests that do not map to a requirement only when they create maintenance risk. If implementation is authorized, make the smallest fix and rerun the affected evidence. Never claim full specification compliance from a green build alone.
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.

