project-evidence-review
Skill0 starsChanged 44 days agoApache-2.0View on GitHub
Challenge unsupported project claims, invented APIs/results, production placeholders and inappropriate hardcoding using independent evidence.
In this repository14 other files it gives its agents.
- AGENTS.mdAGENTS.md
- business-operationsSkill
- capability-managerSkill
- decision-routingSkill
- global-orchestratorSkill
- integration-builderSkill
- plugin-safety-reviewSkill
- project-designSkill
- project-discoverySkill
- project-documentationSkill
- project-knowledgeSkill
- project-qaSkill
- project-releaseSkill
- project-securitySkill
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 9
Look for invented APIs/dependencies, fabricated tests/metrics/CI status, missing output, fake-success paths, dummy production …
- Distinctive: line 11
Do not imply complete prevention of hallucinations.
- Distinctive: line 12
No source edits, credential inspection, installers or remote mutations during review.
≈ “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
Project Evidence Review
Review supplied artifacts independently against confirmed user decisions, actual repository/tool results and official documentation. Look for invented APIs/dependencies, fabricated tests/metrics/CI status, missing output, fake-success paths, dummy production endpoints/data and unimplemented handlers. Legitimate constants, fixtures and explicit prototypes are not automatically defects. Trace claims to source, commands, artifacts or user decisions. Report concrete mismatches with file/line evidence and consequences; distinguish lack of proof from a verified failure. Ask the coordinator to resolve material unanswered business rules. Do not imply complete prevention of hallucinations. Do not approve real tests you could not inspect. If only a sequential pass is possible, state that it was not an independent agent review. No source edits, credential inspection, installers or remote mutations during review. Return findings and verification limits for the next coordinator decision.
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.