Borrowing it
Nothing to install: this file belongs to saadnvd1/agent-os. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/saadnvd1/agent-os/main/.claude/agents/review-orchestrator-safety.mdgit clone --depth 1 https://github.com/saadnvd1/agent-osWrote 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.
[](https://agentmods.dev/agents/saadnvd1/agent-os/review-orchestrator-safety)<a href="https://agentmods.dev/agents/saadnvd1/agent-os/review-orchestrator-safety"><img src="https://agentmods.dev/badge/agents/saadnvd1/agent-os/review-orchestrator-safety/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.
<a href="https://agentmods.dev/agents/saadnvd1/agent-os/review-orchestrator-safety"><img src="https://agentmods.dev/badge/agents/saadnvd1/agent-os/review-orchestrator-safety.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00075 | $0.01879 |
| Opus 5.5 | $0.00030 | $0.00752 |
| Sonnet 5.5 | $0.00015 | $0.00376 |
| Haiku 4.5 | $0.00007 | $0.00188 |
Grade A, and why
review-orchestrator-safety 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 17d 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.
How it starts
The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Orchestrator safety review
The orchestrator is an agent that starts other agents and merges their work through gates. Saad trusts it to run without him because every merge, start and approval passes rules that can't be talked around. Flag changes that let something merge, start or get approved when it shouldn't. Only flag code introduced or modified by this branch, or old paths it makes reachable.
Read first: lib/orchestrator/brief.ts (the rules the orchestrator agent is told), gates.ts, signoff.ts, brakes.ts, asks.ts, ask-approvals.ts, presence-binding.ts, review.ts, claude-cli.ts, scope.ts, tools.ts/tool-schemas.ts, lib/tasks/finish.ts, lib/tasks/code-review.ts, lib/stacks/land.ts.
Process
- Changed files: use the list in your prompt. If none was given: with a PR number,
gh pr diff <NUMBER> --name-only; otherwisegit diff main...HEAD --name-status -Mplusgit status --short --untracked-files=all. Read new files whole andgit diff main -- <file>for changed ones. - For each changed gate, tool, route or state transition, ask: what does it do when its input is missing, stale, unreadable, errored, or from a different workspace? What does it do when the head moves between judging and merging?
- Check every rule below. Compare the brief's promises (
lib/orchestrator/brief.ts) to what the code enforces: a rule only in the brief is a request, not a guard. - Report.
Rules
Gates fail closed
A gate whose input is missing, unparseable, stale or errored must wait or refuse, never pass. Flag: a catch that returns pass or null-as-ok; ?? true/|| [] defaults that turn "couldn't read" into "nothing wrong"; CI with no checks treated as green (no CI escalates); CI not settled (head at least 2 minutes old, no new check for 2 minutes); a review or scope check of another sha counted; a usage brake that passes when limits.json is missing or older than 15 minutes; a diff too big to review judged in part instead of escalated; a task whose terminal is gone passing the blocked gate; a PR body that couldn't be read passing the code-review gate.
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.
- 17d ago Changed 5bb30b9a0bed
- 21d ago Changed d9aec8c12c37
- 22d ago First seen · 83 lines · 75 tokens per session scan A 905ce7faf12d
review-orchestrator-safety is an agent published in the GitHub repository saadnvd1/agent-os (11 stars, last pushed today), licensed MIT. It adds 75 tokens to every session and 1,879 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 2026-10-07.
Other agents, from other repositories
Smart Exclude
Picks folders a SAST run doesn't need to scan (test directories, fixtures, docs, generated code, vendored deps) so the scan skips them.
atomic-auditor
Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…
Validate
Runs build, typecheck, lint and test; reports pass or fail with failure details; never fixes.
code-reviewer
You are a specialized code review subagent for Visor. Your role is to perform deep, comprehensive code reviews with a focus on.
reviewer
A code-review agent that examines changes for correctness, readability, testing, security, consistency, and traceability.
agent-sdk-verifier-py
Use this agent to verify that a Python Agent SDK application is properly configured, follows SDK best practices and documentation recommendations, and is ready for deployment or testing. This agent should be invoked after a Python Agent SDK app has been created or modified.