nimbalyst: Command for Claude Code

.claude/commands/review-contribution.md

review-contribution is a command for Claude Code from nimbalyst/nimbalyst. It costs 22 tokens per session (6,754 once invoked), scanned E, original, MIT.

Review a community-contributed PR for safety, correctness, and adherence to Nimbalyst rules.

Command for Claude Code

Written for Claude Code: installed under .claude/. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions subagents.

This is nimbalyst/nimbalyst's own configuration. It tells Claude Code how to work on nimbalyst itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything nimbalyst configures →

About the project

Nimbalyst is an open-source visual workspace for running coding agents, editing their files and documents, managing tasks, and reviewing changes. It is for developers who want to work with multiple agent sessions, including Codex, Claude Code, and OpenCode, through a desktop application. The catalogue entries are commands, skills, agents, and configuration pieces for its workflows.

nimbalyst/nimbalyst · 1,675 stars · on GitHub · nimbalyst.com

Reuse

Borrowing it

Nothing to install: this file belongs to nimbalyst/nimbalyst. 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/nimbalyst/nimbalyst/main/.claude/commands/review-contribution.md
Clone the repo
git clone --depth 1 https://github.com/nimbalyst/nimbalyst

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.

agentmods badge for review-contribution

README.md
[![agentmods](https://agentmods.dev/badge/commands/nimbalyst/nimbalyst/review-contribution.svg)](https://agentmods.dev/commands/nimbalyst/nimbalyst/review-contribution)
Your own site
<a href="https://agentmods.dev/commands/nimbalyst/nimbalyst/review-contribution"><img src="https://agentmods.dev/badge/commands/nimbalyst/nimbalyst/review-contribution.svg" alt="Measured on agentmods" height="20"></a>
Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 6,754 The whole file, excluding the scripts and references it only reads on demand.
Security scan E 4 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin unknown 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 5.1 $0.00022 $0.06754
Opus 5 $0.00011 $0.03377
Sonnet 5 $0.00004 $0.01351
Haiku 4.5 $0.00002 $0.00675

Measured today against content hash b00472163297, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade E, and why

review-contribution scanned grade E with 4 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 today.

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.

Harvests environment variableshighData exfiltration

Enumerating or grepping the environment for keys collects credentials unrelated to what the mod says it does.

- Reads of credential-bearing state: `process.env` sweeps (`Object.keys(process.env)`), the electron-store `apiKeys` object, `~/.ssh`, `~/.aws`, `~/.npmrc`, keychain APIs, the personal/team JWTs, the PGLite or SQLite dat

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

- Any `curl | sh`, `wget`, or download-and-execute inside a `run:` block.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Any `curl | sh`, `wget`, or download-and-execute inside a `run:` block.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- Any new outbound call -- `fetch`, `axios`, `net.request`, `https.request`, `child_process`, `ws://`/`wss://` -- or a changed base URL, telemetry endpoint, or sync host. Review the *destination*, not just the call site.
.claude/commands/review-contribution.md · 354 lines

How it starts

The opening of the file, as written. The whole thing — 354 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review a community-contributed PR and answer one question: is there anything that blocks merging? Non-blocking improvements get a brief mention so the maintainer can clean them up post-merge -- they do not belong in the response to the contributor.

CRITICAL: Do NOT touch the working directory

This review runs entirely from the GitHub API via gh. The user's current working directory is off-limits.

Forbidden -- never run, even "just to look":

  • gh pr checkout, git checkout, git switch, git restore
  • git fetch, git pull, git remote add, git remote update
  • git apply, git am, git cherry-pick, git merge, git rebase
  • git stash (any form), git reset, git clean, git add
  • Any command that creates or moves refs (including gh pr checkout --detach, git fetch origin pull/N/head)
  • Any write to files in the working tree
  • gh pr merge, gh pr review --approve|--request-changes|--comment, gh pr comment, gh pr edit

Allowed -- read-only only:

  • gh pr view, gh pr diff, gh pr checks, gh pr list, gh api (GET only)
  • git log, git show, git rev-parse, git config --get
  • Read / Grep / Glob against the maintainer's already-checked-out HEAD (useful as "what does this file look like today" context, not as the PR content itself).

If the maintainer has uncommitted changes or is on a feature branch, do not care -- the review does not depend on working-directory state and must not modify it.

Argument

/review-contribution <PR# | PR URL>

A PR identifier is required. If the user runs /review-contribution with no argument, ask for a PR number or URL.

Session phase: planning, never validating

Call update_session_meta once at the start with phase: "planning" and leave it there for the whole review:

update_session_meta({ name: "PR #<number> contribution review", add: ["review", "github-pr"], phase: "planning" })

Reading someone else's code and deciding what to do about it is exploration, and this command writes no code of its own. validating means we made a change and are now testing it -- a session that has edited nothing has nothing to validate. Two specific traps:

Read the full file on GitHub · 354 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. today First seen · 354 lines · 22 tokens per session scan E b00472163297

Subscribe to this mod's changes

review-contribution is a command published in the GitHub repository nimbalyst/nimbalyst (1,675 stars, last pushed today), licensed MIT. It adds 22 tokens to every session and 6,754 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it E with 4 findings (harvests environment variables, downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-08.