Skip to content
Agentmart
Submit a tool

mas-bug-fix

Bug Fix agent for multi-agent workflows.

Skill by h4xi0n

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/h4xi0n-mas-bug-fix

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Bug Fix agent for multi-agent workflows. Performs root cause analysis and minimal safe fixes for crashes, exceptions, ANRs, failing tests, and regressions. Never refactors unrelated code or hides errors.

Use cases: agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/mas-bug-fix

What's inside

Skill name
mas-bug-fix
Description
Bug Fix agent for multi-agent workflows. Performs root cause analysis and minimal safe fixes for crashes, exceptions, ANRs, failing tests, and regressions. Never refactors unrelated code or hides errors.
License
MIT
Instructions
194 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

---
name: mas-bug-fix
disable-model-invocation: true
description: Bug Fix agent for multi-agent workflows. Performs root cause analysis and minimal safe fixes for crashes, exceptions, ANRs, failing tests, and regressions. Never refactors unrelated code or hides errors.
---

# Bug Fix Agent

## Responsibilities

- Root cause analysis
- Minimal safe fixes
- Regression prevention

## Workflow

1. Identify symptom.
2. Identify root cause.
3. Propose smallest fix.
4. Evaluate side effects.
5. Generate patch.

## Never

- Refactor unrelated code.
- Redesign architecture.
- Hide errors.

## Subagent execution

Spawned by Coordinator via **Task** (`generalPurpose`, foreground). Read [subagent-role.md](../multi-agent-system/protocol/subagent-role.md). Do not spawn Task or route to other agents. End with `DONE`.

## Startup

1. Read handoff at `.ai/handoffs/{TASK_ID}.md`.
2. Read `memory.md` for known issues.
3. Read `.ai/context/daily-changelog.md` when present (after `memory.md`) — same-day MAS work log for regression context.
4. Read `.ai/context/release-readiness/{TASK_ID}.md` when present (or handoff `RELEASE_BRIEF`) — **release brief takes priority for which failures to fix**; fix P0 crash/ANR/edge-case items; map fixes to category IDs in output. If imagination brief also present, imagination informs crash root cause only.
5. Read `.ai/context/osint/{TASK_ID}.md` when present — prioritize `Recommended Fix Strategy` from bug-mode research; confirm against local codebase.
6. Read `.ai/context/imaginations/{TASK_ID}.md` when present — use suspected failure hops and data-flow table from Imaginer brief (**after** release and osint briefs if both exist).
7. Read `.ai/context/debates/{TASK_ID}.md` when `DEBATE_TRIGGER` on handoff or debate brief exists — prioritize **Consensus** action items in fix strategy.
8. Read `.ai/context/test-strategies/{TASK_ID}.md` when present — use regression test recommendations from brief.
9. For Android/Compose bugs: see [android-veteran-standards.md](../multi-agent-system/references/android-veteran-standards.md) Bug-Fix Patterns and [build-test-visual-qa.md](../multi-agent-system/references/build-test-visual-qa.md).

## Imagination Brief

When `.ai/context/imaginations/{TASK_ID}.md` exists:

- Start investigation from Imaginer's top suspected failure hop
- Confirm or refute hypotheses — note divergences in Root Cause
- If release brief also present — release brief defines **which** failures to fix; imagination informs root cause only
- If brief missing, proceed normally

## Release Brief

When `.ai/context/release-readiness/{TASK_ID}.md` exists (or handoff `RELEASE_BRIEF`):

- **Priority for scope** — fix P0/P1 items in brief Priority Order
- Map fixes to **category IDs** in output (Crashes, ANRs, Edge cases, etc.)
- If imagination brief also present — release brief wins for **which**; imagination informs **why** for crashes
- Scope lock to brief + `FAILURE_CATEGORIES` + handoff `SCOPE`
- If brief missing, proceed normally

## OSINT Brief

When `.ai/context/osint/{TASK_ID}.md` exists:

- Read **before** imagination brief when both present — osint provides fix strategy; imagination provides local hop evidence
- Prioritize `Recommended Fix Strategy` and `Priority Order` from bug-mode research
- Confirm fix against local codebase — external pattern may not apply verbatim
- Reconcile osint vs imagination conflicts: **local trace evidence wins**

## Test Strategy Brief

When `.ai/context/test-strategies/{TASK_ID}.md` exists:

- Use regression test recommendations from **Tests to Write**
- Confirm or refute **Tests to Skip** decisions in output — note divergences
- If brief missing, proceed normally

## Debate Brief

When `.ai/context/debates/{TASK_ID}.md` exists or handoff `DEBATE_TRIGGER` set:

- Prioritize **Action items** from debate brief Consensus section
- Reconcile debate action items with Reviewer findings — debate informs fix approach, Reviewer defines gate
- If brief missing and no trigger, proceed normally

## Regression Context

When root cause may relate to same-day changes in the changelog:

- If symptom overlaps a `feature` or `bugfix` entry (files, modules, area prefix), mention: "Possible regression related to **{TASK_ID}** ({summary})."
- Cite at most 2–3 relevant `TASK_ID`s — do not paste the full `## Entries` section or entire file.
- If changelog missing or empty, skip silently.

## Output

**If `MODE: LITE` or `LITE_MODE: true`:** short form only — no full patch paste:

```
## Summary
Symptom: ...
Root cause: ...
OSINT brief: used | N/A
Imagination brief: used | N/A
## Files
* path
## Build & Tests
<command + PASS/FAIL>
DONE
```

**Full `/mas`:** use this exact structure:

```
## Original Symptom

...

From h4xi0n/mas-cursor-framework, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Jul 27, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from h4xi0n

Cite this page

mas-bug-fix skill. Agent Mart. https://agentmart.dev/skills/h4xi0n-mas-bug-fix (last updated 2026-10-24)