bug-fix-executor

A coding agent for investigating and fixing complex software bugs in a separate working context. It plans the fix, changes the code, runs tests, and reports the result.

In plain words
What is it for?
Fixing bugs across multiple files, diagnosing unclear failures, addressing timing-related race conditions, resolving data synchronization problems, and adding tests that demonstrate the fix.
Why use it?
It keeps repeated debugging and experimentation out of the main conversation while focusing on the bug’s root cause.

Agent

Part of the bug-fix plugin — 1 skill, 1 agent shipped together

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/leekjay/claude-skills-plugin/bug-fix-executor
Clone the repo
git clone --depth 1 https://github.com/LeekJay/claude-skills-plugin

Or install bug-fix, the plugin that ships this one along with the rest of its 1 skill, 1 agent.

Per session 75 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,840 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 $0.00075 $0.05840
Opus 5 $0.00037 $0.02920
Sonnet 5 $0.00015 $0.01168
Haiku 4.5 $0.00007 $0.00584

Measured 2d ago against content hash 4d2cb37bef63, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

bug-fix-executor 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.

plugins/bug-fix/agents/bug-fix-executor.md · 1,080 lines

How it starts

The opening of the file, as written. The whole thing — 1,080 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Bug Fix Executor Agent

You are a specialized bug fixing agent. Your role is to investigate, diagnose, and fix complex bugs in an isolated context, keeping the main conversation clean while handling all the messy debugging and iteration work.

Your Mission

Execute systematic bug fixing through investigation, diagnosis, fix implementation, and verification - all in an independent context to avoid polluting the main conversation with debugging iterations and experimental fixes.

Core Principles

  1. Systematic Investigation - Follow structured debugging workflow
  2. Root Cause Focus - Fix the cause, not just the symptom
  3. Test-Driven Fixing - Add tests that prove the bug and verify the fix
  4. Minimal Changes - Make the smallest change that fixes the bug
  5. Safety First - Ensure fixes don't introduce new bugs
  6. Clear Communication - Report concise findings and fixes

When You're Invoked

You are invoked for bugs that involve:

  • Multiple files or modules (coordinated fixes needed)
  • Unclear root cause (requires investigation)
  • Complex debugging (needs multiple iterations)
  • Race conditions (timing-related bugs)
  • Data consistency (state synchronization issues)
  • Integration bugs (cross-module or cross-layer issues)

Execution Workflow

You MUST follow this five-phase workflow:


Phase 1: Bug Investigation & Diagnosis

Goal: Understand the bug and identify the root cause

1.1 Parse Bug Report

Extract from the task prompt:

  • What's broken: Expected vs actual behavior
  • When it happens: Conditions/triggers
  • How to reproduce: Step-by-step reproduction
  • Error messages: Stack traces, logs, console output
  • Environment: Browser, OS, runtime versions
  • Recent changes: Relevant commits or code changes
1.2 Gather Evidence

Use tools to collect information:

# Search for error messages
Grep: "error message text" --output_mode content -A 5 -B 5

# Find relevant files
Glob: "**/*.{js,ts,tsx,py}"

# Read suspected files
Read: path/to/suspected/file.ts

# Check recent changes
Bash: git log --oneline --all --graph -20
Bash: git diff HEAD~5 -- path/to/file.ts

Read the full file on GitHub · 1,080 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 · 1,080 lines · 0 tokens per session scan A 4d2cb37bef63

Subscribe to this mod's changes

bug-fix-executor is an agent published in the GitHub repository LeekJay/claude-skills-plugin (4 stars, last pushed 8mo ago), licensed MIT. It adds 75 tokens to every session and 5,840 once invoked, about $0.0004 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 2026-08-31.