evidence-driven-debugging
Skill2 starsChanged 12 days agoMITView on GitHub
Investigate and fix defects by reproducing the reported path, comparing hypotheses, and checking runtime behavior. Use for bugs, regressions, intermittent failures, and fixes needing evidence beyond a plausible code change.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- ChloeVPin/open-agent-skillstheir own version, first seen Aug 30, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- ChloeVPin/open-agent-skills★ 2same 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 repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- aeoSkill
- agent-worktree-orchestrationSkill
- critical-adversarial-reviewSkill
- gauntlet-loopSkill
- git-quality-hooksSkill
- multi-agent-collaborationSkill
- quality-gatesSkill
- source-backed-researchSkill
- spec-to-planSkill
- validated-coding-loopSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Follow the exact path the user reported.
- Distinctive: line 30
If reproduction remains unavailable, distinguish a proposed fix from a verified correction.
≈ “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
Evidence-driven debugging
Follow the exact path the user reported. A passing check on a nearby subsystem does not explain the defect.
Find a decisive observation
Record the symptom, expected behavior, revision, inputs, and environment. Trace the action through the real UI, API, persisted state, or artifact pipeline.
Reproduce the failure or identify the smallest observation separating the leading explanations. Include a competing explanation for an obvious first guess when evidence is ambiguous.
A page can display stale data while the query is correct. An asset can exist while production imports an older hash. A count can measure events while the user asked about distinct participants.
Make the check useful
Choose the smallest check exercising the failing behavior. If adding a regression gate, show that it detects the broken state before fixing it. A deliberate negative control can reveal whether the gate reads runtime behavior.
A screenshot, network response, or raw measurement can be decisive. Use the form appropriate to the symptom. Do not replace a visible failure with a claim based only on syntax or a build.
Fix and verify
Change the cause within scope. Rerun the failure case and nearby affected contracts. Preserve unrelated work, ownership boundaries, and accepted product decisions.
Record the cause, changed behavior, verification command or capture, and material limits in the project's designated bug or continuation record.
If reproduction remains unavailable, distinguish a proposed fix from a verified correction. Timeouts and incomplete measurements remain unknown.
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.