Run any Skill in Manus with one click

fix-bug

Stars83

Forks14

UpdatedJune 10, 2026 at 09:15

Three-phase autonomous bug fix — investigate all occurrences, fix with full coverage, validate with regression test. Prevents partial fixes (the

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Run Skill in Manus

Source

escapeboy

escapeboy/ai-prompts

View GitHub Repository View Creator Repositories

Download

Run Skill in Manus

Related occupationsSOC

Based on SOC occupation classification

Software DevelopersComputer and Mathematical Occupations·SOC 15-1252

SKILL.md

readonly

name	fix-bug
description	Three-phase autonomous bug fix — investigate all occurrences, fix with full coverage, validate with regression test. Prevents partial fixes (the

fix-bug

Three-phase bug fix that finds ALL occurrences, fixes atomically, and validates with a regression test.

Phase 1 — Comprehensive Investigation

Before touching any code:

Reproduce the issue and capture the exact error message / stack trace
Identify the primary symbol involved (function name, route, variable, class, column)
Grep the entire codebase for every reference to that symbol:
- Controllers, services, models
- View templates (Blade / JSX / ERB / Twig — server-rendered views are the most-missed spot)
- Tests (feature + unit)
- Route definitions (all route files, not just the obvious one)
- Config files
- JS/CSS files if relevant
Document: full list of files that reference the broken symbol
Check for the same bug pattern in similar locations (e.g. if one route is broken, check all routes of the same type)

Do not proceed until you have the complete reference map.

Phase 2 — Fix with Full Coverage

Apply changes atomically across every file found in Phase 1:

For renamed variables/properties: update the definition + every view, controller, test, and config
For broken routes: check all route files, all route-helper calls, all redirects and links
For removed/changed model attributes: update mass-assignment lists ($fillable or equivalent), factories, seeders, and all tests that reference those attributes
For middleware changes: check route groups, kernel/middleware registration, and all test middleware overrides

After applying all changes, grep again to confirm zero remaining old references.

Phase 3 — Validate

Run the full test suite (php artisan test --parallel, npm test, pytest — project equivalent)
If tests fail:
- Determine: is the test wrong, or is the fix incomplete?
- Never change test assertions to make them pass trivially
- Fix the underlying issue
Write a regression test that:
- Reproduces the original bug (assert it was broken)
- Confirms the fix works
- Lives in the appropriate test directory
Final grep: confirm zero references to the old broken pattern

Output

Report back:

Root cause (1-2 sentences)
Files changed (list)
Regression test location
Test suite result (pass/fail count)

The bug to fix: $ARGUMENTS

More from this repository

same repository

cache-inspector

escapeboy/ai-prompts

Analyze prompt cache hit rates, estimate cost savings from cached system prompts and memories, and recommend improvements to caching strategy. Use when checking cache performance, investigating high token costs, optimizing cache hit rates, or diagnosing slow cache warmup.

2026-07-1883

context

escapeboy/ai-prompts

Manage Serena memories and session context — load, save, refresh, inspect, and clear project knowledge

2026-07-1883

init-project

escapeboy/ai-prompts

Initialize Claude Code optimization for a new project — detect stack, create memories, generate constitution, configure settings

2026-07-1883

optimize

escapeboy/ai-prompts

Reduce token usage by selecting a planning strategy, loading Serena memories, enforcing symbol-first exploration, and reporting savings. Use when starting a non-trivial task, asking to reduce token costs, requesting efficient execution, or wanting optimized Claude Code workflows.

2026-07-1883

update-docs

escapeboy/ai-prompts

Search the web for latest Claude API changes, compare findings to existing documentation, and apply targeted updates to keep prompts and configs current. Use when docs are outdated, after a Claude API update, to refresh system prompts, or to validate documentation accuracy.

2026-07-1883

self-improve

escapeboy/ai-prompts

Close the loop on a skill/prompt/convention library: mine recurring review feedback into rules, fold them back into the generator, and gate every change through a three-tier evaluation (deterministic lint + trigger accuracy + LLM judge) that converges instead of drifting. Use when a correction keeps recurring across skills/PRs, when adding or revising skills warrants a quality gate, or when asked to "harden the skill library", "run the eval", "lint skills", or "close the loop". Composes with your existing execution, memory, and governance layers.

2026-07-1883

fix-bug

Three-phase bug fix that finds ALL occurrences, fixes atomically, and validates with a regression test.

Phase 1 — Comprehensive Investigation

Before touching any code:

Reproduce the issue and capture the exact error message / stack trace

Identify the primary symbol involved (function name, route, variable, class, column)

Grep the entire codebase for every reference to that symbol:

Controllers, services, models
View templates (Blade / JSX / ERB / Twig — server-rendered views are the most-missed spot)
Tests (feature + unit)
Route definitions (all route files, not just the obvious one)
Config files
JS/CSS files if relevant

Document: full list of files that reference the broken symbol

Check for the same bug pattern in similar locations (e.g. if one route is broken, check all routes of the same type)

Do not proceed until you have the complete reference map.

Phase 2 — Fix with Full Coverage

Apply changes atomically across every file found in Phase 1:

For renamed variables/properties: update the definition + every view, controller, test, and config

For broken routes: check all route files, all route-helper calls, all redirects and links

For removed/changed model attributes: update mass-assignment lists ($fillable or equivalent), factories, seeders, and all tests that reference those attributes

For middleware changes: check route groups, kernel/middleware registration, and all test middleware overrides

After applying all changes, grep again to confirm zero remaining old references.

Phase 3 — Validate

Run the full test suite (php artisan test --parallel, npm test, pytest — project equivalent)

If tests fail:

Determine: is the test wrong, or is the fix incomplete?
Never change test assertions to make them pass trivially
Fix the underlying issue

Write a regression test that:

Reproduces the original bug (assert it was broken)
Confirms the fix works
Lives in the appropriate test directory

Final grep: confirm zero references to the old broken pattern

Output

Report back:

Root cause (1-2 sentences)

Files changed (list)

Regression test location

Test suite result (pass/fail count)

The bug to fix: $ARGUMENTS