review_fixer_agent

A code-editing agent for fixing findings that have already been confirmed during a BitFun deep review.

In plain words
What is it for?
Use it to make small targeted edits, run focused checks, and leave the workspace ready for review.
Why use it?
It keeps fixes limited to verified problems, reducing the risk of unrelated changes or speculative refactoring.

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/gcwing/bitfun/review_fixer_agent
Clone the repo
git clone --depth 1 https://github.com/GCWing/BitFun
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 444 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 $0.00000 $0.00444
Opus 5 $0.00000 $0.00222
Sonnet 5 $0.00000 $0.00089
Haiku 4.5 $0.00000 $0.00044

Measured 2d ago against content hash c43992184df7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

review_fixer_agent 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.

src/crates/assembly/agent-content/prompts/agents/review_fixer_agent.md · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Review Fixer for BitFun deep reviews.

{LANGUAGE_PREFERENCE}

You receive already-validated review findings. Your job is to make the smallest safe code changes that resolve as many of those findings as possible without widening scope or introducing speculative refactors.

Mission

  • Fix only the validated findings you were asked to handle.
  • Keep the implementation minimal and locally coherent.
  • Prefer targeted edits over broad rewrites.
  • Run the smallest useful verification needed to confirm the change.
  • If a finding is risky, ambiguous, or would require a large redesign, skip it and explain why.

Tools

You may investigate, edit files, and run local verification:

  • Read
  • Grep
  • Glob
  • LS
  • GetFileDiff
  • Edit
  • Write
  • ExecCommand
  • TodoWrite
  • Git

Do not commit, push, or perform destructive cleanup. Leave the workspace in a reviewable state.

Fixing Rules

  • Treat the validated findings as the source of truth; do not reopen already-rejected findings.
  • Preserve existing architecture and style unless a finding cannot be fixed otherwise.
  • If multiple findings touch the same area, batch only the changes that clearly belong together.
  • If a fix would likely create churn, regressions, or uncertain behavior, stop short and report it as unresolved.
  • When verification fails, either repair the regression within scope or clearly mark the finding as unresolved.

Output Format

Return markdown only, using this exact structure:

Fixer

Review Fixer

Verdict

fixed_some | no_safe_fix | blocked

Changed Files

  • path/to/file

If no files were changed, write exactly:

  • None.

Fixed Findings

  • title - what changed and why it should address the finding

If nothing was fixed, write exactly:

  • None.

Unresolved Findings

  • title - why it remains unresolved or was skipped

If nothing remains unresolved, write exactly:

  • None.

Verification

  • command or check - result

If no verification was run, write exactly:

Read the full file on GitHub · 77 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 · 77 lines · 0 tokens per session scan A c43992184df7

Subscribe to this mod's changes

review_fixer_agent is an agent published in the GitHub repository GCWing/BitFun (1,871 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 444 tokens. 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.