Agent skill · Documentation

research-after-failure

Use after 2 consecutive failed attempts at solving a problem - STOP guessing and research documentation, codebase, and online resources before resuming

majiayu000github.com/majiayu000GitHub ↗
claude-coderead-onlyMIT
Install
npx skills add majiayu000/claude-skill-registry --skill research-after-failure-troykelly-claude-skills --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 2
SKILL.md size: 6 KB
Bundled scripts: none
Allowed tools: -Read-Grep-Glob-WebFetch-WebSearch
Path: skills/analysis/research-after-failure-troykelly-claude-skills/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 534
Language: HTML

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

From the SKILL.md

# Research After Failure ## Overview After 2 failed attempts, stop and research. Don't keep trying the same thing. **Core principle:** Insanity is doing the same thing and expecting different results. **Trigger:** Two consecutive failed attempts at solving a problem. **Announce at start:** "I've failed twice. I'm stopping to research before trying again." ## The Rule ``` Attempt 1: Try solution │ ▼ Failed? │ ┌──────┴──────┐ │ │ Yes No → Done │ ▼ Attempt 2: Try different approach │ ▼ Failed? │ ┌──────┴──────┐ │ │ Yes No → Done │ ▼ STOP │ ▼ RESEARCH ← You are here │ ▼ Attempt 3: Try with new knowledge ``` ## What Counts as a Failure | Failure | Not a Failure | |---------|---------------| | Tests don't pass | Minor syntax error fixed | | Build breaks | Typo corrected | | Feature doesn't work | IDE autocomplete issue | | Same error recurs | Different error (progress) | | No progress made | Partial progress | ## The Research Protocol ### Step 1: Document the Failures Before researching, document what was tried: ```markdown ## Failed Attempts ### Attempt 1 **Approach:** [What was tried] **Result:** [What happened] **Error:** [Error message if any] ### Attempt 2 **Approach:** [What was tr

What's inside
Steps it walks through
  1. Overview
  2. The Rule
  3. What Counts as a Failure
  4. The Research Protocol
  5. Step 1: Document the Failures
  6. Step 2: Research Repository Documentation
  7. Step 3: Research Existing Codebase
  8. Step 4: Research Online
  9. Step 5: Synthesize Findings
  10. Step 6: Update Issue
  11. Step 7: Resume with New Knowledge
  12. When to Escalate
  13. Research Anti-Patterns
  14. What to Research First
Ships with 1 file
  • metadata.json
Commands it runs
Check README
cat README.md
Check docs directory
ls -la docs/
cat docs/[relevant-topic].md
Check CONTRIBUTING
cat CONTRIBUTING.md
Search for relevant docs
grep -r "[keyword]" docs/
Find similar patterns
More from claude-skill-registry
All skills →
About this skill
What does the research-after-failure skill do?

Use after 2 consecutive failed attempts at solving a problem - STOP guessing and research documentation, codebase, and online resources before resuming

How do I install it?

Run `npx skills add majiayu000/claude-skill-registry --skill research-after-failure-troykelly-claude-skills --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going