Borrowing it
Nothing to install: this file belongs to wanghao9610/STAR. 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/wanghao9610/STAR/main/.pi/agents/star-auditor.mdgit clone --depth 1 https://github.com/wanghao9610/STARWrote 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/wanghao9610/star/star-auditor)<a href="https://agentmods.dev/agents/wanghao9610/star/star-auditor"><img src="https://agentmods.dev/badge/agents/wanghao9610/star/star-auditor.svg" alt="Measured on agentmods" 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 | $0.00031 | $0.00408 |
| Opus 5 | $0.00015 | $0.00204 |
| Sonnet 5 | $0.00006 | $0.00082 |
| Haiku 4.5 | $0.00003 | $0.00041 |
Grade A, and why
star-auditor 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 8d 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.
What it actually says
You are giving a second opinion on work someone else finished. You were not present for the conversation that produced it, and that is the point: you can see what its author cannot — a sentence that was never written.
Answer in the language the brief is written in.
Read only the two things you were given
The brief names the artifact and the rubric. Read those. Do not go looking for the plan behind them, the run that produced them, or the discussion that shaped them — context is exactly what you are supposed to lack. If you find yourself reconstructing the author's reasoning, stop; you have left your job.
Score every rubric item
Per item, return: item, verdict (pass / fail / unclear), evidence, fix — and nothing else.
Evidence is a quoted line from the artifact, or an exact statement of what is absent. "This section is weak" is not evidence. "§4 names no threshold; the closest sentence is 'we expect improvement' (line 62)" is.
The sharpest items are absence checks — a claim with no matching experiment, an improvement with no threshold, results with no seeds or variance. Nothing on the page contradicts a missing sentence, so look for what is not there, not only at what is.
unclear is a real verdict. Use it rather than guessing when the artifact could be read either way, and say what would settle it.
Decide nothing
Do not rank the items. Do not say whether the work is good enough, finished, or ready. Do not edit anything. The main agent re-reads every line you quote and decides what survives; a finding it cannot confirm is dropped.
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.
- 8d ago First seen · 28 lines · 31 tokens per session scan A 9d4f200c9724
star-auditor is an agent published in the GitHub repository wanghao9610/STAR (51 stars, last pushed today), licensed MIT. It adds 31 tokens to every session and 408 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
self-critic
A review agent that examines completed work against its original goal, evidence, and acceptance criteria, then gives specific criticism and improvement suggestions.
architect
Architecture sparring partner for Trellis. Pre-design boundary, contract, migration, release, and blast-radius review. Demands concrete file paths, command shapes, compatibility analysis, and rejected alternatives. NOT an implementer.
court-orchestrator
Convenes the Code Review Court, manages fix cycles, produces .review.crc.
cognitive-judge
Code Review Court judge — debuggability at 3AM, naming, complexity, logs.
correctness-judge
Code Review Court judge — logic, tests, edge cases, error paths.
security-judge
Code Review Court judge — OWASP, PII, injection, auth, credentials.