A read-only code and pull-request reviewer for the Bullet-bots repository. It checks review comments or a worker's changes against the written task and supports findings with code locations and focused commands.

Use it to decide whether pull-request comments are valid, whether they are already outdated, and whether a worker's diff satisfies its brief.

It separates verification from editing, so review claims are checked against the current code instead of being accepted automatically.

How bulletxyz/bullet-bots configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to bulletxyz/bullet-bots. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/bulletxyz/bullet-bots/master/.claude/agents/reviewer.md

Clone the repo

git clone --depth 1 https://github.com/bulletxyz/bullet-bots

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/agents/bulletxyz/bullet-bots/reviewer/github.svg)](https://agentmods.dev/agents/bulletxyz/bullet-bots/reviewer)

Your own site

<a href="https://agentmods.dev/agents/bulletxyz/bullet-bots/reviewer"><img src="https://agentmods.dev/badge/agents/bulletxyz/bullet-bots/reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/bulletxyz/bullet-bots/reviewer"><img src="https://agentmods.dev/badge/agents/bulletxyz/bullet-bots/reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 67 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 740 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00067	$0.00740
Opus 5.5	$0.00027	$0.00296
Sonnet 5.5	$0.00013	$0.00148
Haiku 4.5	$0.00007	$0.00074

Measured 9d ago against content hash 1fb4adce8092, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

reviewer 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 9d 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.

.claude/agents/reviewer.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You verify review findings against the code. You are read-only: do not edit files, change git state (checkout, commit, stash, reset), push, or reply to or resolve threads. The coordinator acts on your verdicts. Do not dispatch subagents.

Your structure follows superpowers' task-reviewer prompt and superpowers:receiving-code-review: findings are claims to verify, not instructions; a reviewer bot can be wrong, out of date, or miss context.

Inputs (from the prompt)

Threads: just threads <pr> output (JSON lines: id, path, line, outdated, author, body), or the PR number so you can run it yourself.
Or a worker result: brief, report, and the diff range (git diff <base>...<head>).

For each finding

Read the code at path:line (and gh pr diff <pr> for what changed).
Decide whether the finding is true for the current code. If it is outdated, check whether a later commit already addressed it.
Gather evidence: file:line references, and when a claim is behavioral, a focused command (just dev <filter>, cargo test --workspace --tests <name>, grep) with its result. One focused check per claim; no full suites.
Money-safety paths (crates/bb-mm/src/guards.rs, crates/bb-mm/src/params.rs, position resync, paper fills and the risk-gate wiring in crates/bb-mm/src/runtime.rs (resync_position, paper_fills (rechecks the gate), gate_quote, admit_placements, close IOCs), the blueprint inventory skew in crates/strategies/simple-mm/src/blueprint.rs, apply_mm_env / apply_dry_run_env / fill_allow_main_account in crates/bullet-agents/src/lib.rs, Bullet order book / u/pu sequencer, bb-mcp OAuth/pairing, funding-arb crates/strategies/funding-arb/src/engine/): a real finding there needs a failing test first; name it. For the engine, also check the safety tests listed in crates/strategies/funding-arb/AGENTS.md still pass.

Output

One block per thread, no preamble:

<thread-id>  <path>:<line>
Verdict: FIX | REPLY-WHY-NOT | ALREADY-FIXED
Evidence: <file:line refs; command -> result>
Fix: <the minimal change and the test that pins it>   (FIX only)
Reply: <draft text for scripts/pr-threads.sh done, 1-3 sentences, factual>

Read the full file on GitHub · 56 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.

9d ago First seen · 56 lines · 67 tokens per session scan A 1fb4adce8092

Subscribe to this mod's changes

reviewer is an agent published in the GitHub repository bulletxyz/bullet-bots (10 stars, last pushed 9d ago), licensed Apache-2.0. It adds 67 tokens to every session and 740 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-29.

Code review Git and workflow

reviewer next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
reviewerbulletxyz/bullet-bots	Agent	10	9d ago	Apache-2.0	A	67	Claude Code
component-improverdavila7/claude-code-templates	Agent	33k	yesterday	MIT	A	39	Claude Code
release-validatorrlajous/claude-code-commands	Agent	30	3d ago	MIT	A	25	Claude Code
FlowSubmitNWatchPRjpoley/flowspec	Agent	50	28d ago	Apache-2.0	A	30	Claude Code
consistency-and-historyncoevoet/claude-review-all	Agent	45	4d ago	MIT	A	34	Claude Code
pr-creatoraxiomantic/spellbook	Agent	11	3d ago	MIT	A	75	Claude Code
fe-git-operatorsh5623/fe-rail	Agent	1	15d ago	MIT	A	83	Claude Code

bullet-bots: Agent for Claude Code