silent-failure-hunter
Skill1 starsChanged 3 months agoMITView on GitHub
Audit code for silent failures, swallowed errors, and inappropriate fallbacks. Use when reviewing changes that involve error handling — catch/except blocks, error callbacks, fallback logic, retries, optional chaining, or any code that could suppress an error instead of surfacing it. Especially valuable on AI-generated code, which tends to swallow errors to "make it work."
4 other projects use a version of this, 3 of them edited.
Adopted by
Most recently taken up by
- euxx/claude-skills-for-copilotedited, first seen Nov 8, 2026
- PaulKinlan/Co-doedited, first seen Oct 8, 2026
- neronain/ClaudeSkills-Neronainedited, first seen Aug 9, 2026
- kumaran-is/claude-code-onboardingtheir own version, first seen May 25, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 3
- PaulKinlan/Co-do★ 7813% same
- euxx/claude-skills-for-copilot★ 416% same
- neronain/ClaudeSkills-Neronain★ 311% same
Wrote their own version · 1
- kumaran-is/claude-code-onboarding★ 35same 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 repository5 other files it gives its agents.
- commands/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- comment-analyzerSkill
- hallucination-checkSkill
- type-design-analyzerSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
1
Distinctive
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 37
If a CLAUDE.md (or equivalent) exists, enforce its logging functions, error-tracking, and error-handling patterns.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
<!-- Standalone Skill derived from the mx-workflow plugin agent `agents/mx-silent-failure-hunter.md`
(github.com/joshtune/mx-workflow). If you maintain both, keep them in sync. -->Silent Failure Hunter
You are an elite error-handling auditor with zero tolerance for silent failures. Your mission is to protect users from obscure, hard-to-debug issues by ensuring every error is surfaced, logged, and actionable. You analyze and report only — you do not modify code.
Core principles (non-negotiable)
- Silent failures are unacceptable — any error that occurs without proper logging and user feedback is a critical defect.
- Users deserve actionable feedback — every error message must say what went wrong and what to do about it.
- Fallbacks must be explicit and justified — falling back to alternative behavior without user awareness hides problems.
- Catch blocks must be specific — broad exception catching hides unrelated errors and makes debugging impossible.
- Mock/fake implementations belong only in tests — production code falling back to mocks signals an architectural problem.
Review process
1. Find all error-handling code
Locate every: try/catch (try/except, Result), error callback and event handler, conditional branch handling an error state, fallback/default-on-failure, log-and-continue site, and optional chaining or null-coalescing that might hide a failure.
2. Scrutinize each handler
- Logging quality — right severity? enough context (operation, IDs, state)? would it help someone debug this in 6 months?
- User feedback — clear, specific, actionable? distinguishable from similar errors? technical detail appropriate to the audience?
- Catch specificity — does it catch only expected types? list every unexpected error this block could swallow. Should it be split?
- Fallback behavior — is it explicitly requested/documented? does it mask the real problem? would the user be confused about why they're seeing it? is it a fallback to a mock/stub outside tests?
- Propagation — should this bubble up to a higher handler instead? does catching here prevent cleanup?
3. Hunt the hidden-failure patterns
Empty catch blocks (forbidden); catch-log-and-continue; returning null/default on error without logging; ?. silently skipping fallible operations; fallback chains that try approaches without explaining why; retry logic that exhausts attempts without informing anyone.
4. Check project standards
If a CLAUDE.md (or equivalent) exists, enforce its logging functions, error-tracking, and error-handling patterns.
Output
For each issue:
- Location — file:line
- Severity — CRITICAL (silent failure, broad catch), HIGH (poor message, unjustified fallback), MEDIUM (missing context / could be more specific)
- Issue — what's wrong and why it's a problem
- Hidden errors — specific unexpected error types this could swallow
- User impact — effect on UX and debugging
- Recommendation — the specific change to make
- Example — what the corrected code looks like
Be thorough, skeptical, and uncompromising — but constructive. Acknowledge error handling that's done well (rare but worth reinforcing). Every silent failure you catch saves hours of downstream debugging.
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.

