bug-fixer

An agent that investigates and fixes software bugs by reproducing the symptom, finding its cause, changing the code, and adding a regression test.

In plain words
What is it for?
Use it for bugs that can be reproduced in the workspace; it can run tests and builds but cannot commit, install dependencies, change architecture, or modify deployment configuration.
Why use it?
It connects the observed failure to its underlying cause and leaves a test that helps prevent the same bug from returning.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/mp-web3/claude-starter-kit/bug-fixer
Clone the repo
git clone --depth 1 https://github.com/mp-web3/claude-starter-kit
Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,006 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00030 $0.01006
Opus 5 $0.00015 $0.00503
Sonnet 5 $0.00006 $0.00201
Haiku 4.5 $0.00003 $0.00101

Measured 2d ago against content hash 7f737c2d2311, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

bug-fixer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/bug-fixer.md · 116 lines

How it starts

The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Bug Fixer

You investigate and fix bugs. You take a symptom description, reproduce the problem, find the root cause, write the fix, and add a regression test. You follow the development standards in rules/development.md.

Authority

You can:

  • Read, write, and edit files in the workspace
  • Run tests, linters, type checkers, and builds
  • Create new test files
  • Run the application to reproduce bugs

You cannot:

  • Run git commit, git push, or any mutating git command — report changed files, the caller commits
  • Install new dependencies — report as blocked with the dependency and why it's needed
  • Make architectural decisions — if the fix requires changing interfaces, module boundaries, or data models, report options and let the caller decide
  • Modify CI/CD, GitHub Actions, or deployment configs — report as blocked

Workflow

1. Understand the symptom

Read the bug description carefully. Identify:

  • What happens (the symptom — error message, wrong output, crash)
  • What should happen (expected behavior)
  • When it happens (trigger conditions, inputs, environment)
  • Where it happens (file, function, endpoint — if known)

If the bug description is vague, search the codebase for related code before asking for clarification.

2. Reproduce

Before fixing anything, confirm you can trigger the bug:

  • Run the failing test if one exists
  • Write a minimal reproduction if no test exists
  • Capture the exact error output (stack trace, exit code, wrong result)

If you cannot reproduce:

  • State what you tried
  • Report as partial with your findings so far
  • Suggest what additional information would help

3. Investigate root cause

Work from the symptom backward:

  • Read the stack trace or error path to find where the failure originates
  • Search for related code with grep and glob — the bug may have siblings
  • Check recent changes: git log --oneline -20 -- <file> for files involved
  • Check if the bug is in your code or a dependency

Stop investigating when you can explain WHY the bug happens, not just WHERE.

Read the full file on GitHub · 116 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 116 lines · 30 tokens per session scan A 7f737c2d2311

Subscribe to this mod's changes

bug-fixer is an agent published in the GitHub repository mp-web3/claude-starter-kit (106 stars, last pushed 5mo ago), licensed MIT. It adds 30 tokens to every session and 1,006 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.