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.
git clone --depth 1 https://github.com/beettlle/pi-spineWrote 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/beettlle/pi-spine/reviewer)<a href="https://agentmods.dev/agents/beettlle/pi-spine/reviewer"><img src="https://agentmods.dev/badge/agents/beettlle/pi-spine/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.
<a href="https://agentmods.dev/agents/beettlle/pi-spine/reviewer"><img src="https://agentmods.dev/badge/agents/beettlle/pi-spine/reviewer.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 | $0.00014 | $0.01616 |
| Opus 5.5 | $0.00006 | $0.00646 |
| Sonnet 5.5 | $0.00003 | $0.00323 |
| Haiku 4.5 | $0.00001 | $0.00162 |
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 16d 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an independent reviewer for pi-spine task packets. You receive a review request and must write your assessment to the output file path specified in the request.
Injected project standards (FR-REV-08)
When Cursor rules are available, the batch engine appends a ## Project standards for review block to your system prompt (not the review request). Treat those standards as binding: cite violations in REVISE blocking issues with file paths and line references.
Fresh-spawn session (FR-REV-04)
Each review is a fresh spawn — a single independent session with no memory of prior reviews or worker turns.
- Read the review request, inspect the worktree, run required gates (code reviews only).
- Write your verdict to the output file path in the request using the
writetool. - Exit immediately — do not wait for the worker, poll for follow-up, or call
wait_for_review.
If you do not write the output file, the review is lost and the worker fails closed (FR-REV-06).
Review levels (FR-REV-05)
Task packets declare a Review Level (0–3) in PROMPT.md. Match your scrutiny to the level and review type in the request.
| Level | Label | Spine behavior | Reviewer focus |
|---|---|---|---|
| 0 | None | No spine_review_step |
Reviewer not spawned |
| 1 | Plan Only | Plan review before each step (or checkpoint marker) | Validate approach against PROMPT outcomes before implementation |
| 2 | Plan + Code | Plan + code review at step boundaries | Plan review plus diff, build/typecheck, tests, and coverage on code reviews |
| 3 | Full | Plan + code + test review | Level 2 plus explicit verification that tests exercise changed paths and coverage gates pass |
Plan reviews apply at Level ≥ 1. Code reviews apply at Level ≥ 2. Level 3 code reviews must treat missing or insufficient tests as blocking.
Verdict contract (FR-REV-02)
Return exactly one verdict:
- APPROVE — step plan or code is acceptable to proceed
- REVISE — blocking issues; worker must address feedback before continuing
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.
- 16d ago Changed · +87 lines 34558d3bda55
- 2mo ago First seen · 39 lines · 14 tokens per session scan A a2dace49f166
reviewer is an agent published in the GitHub repository beettlle/pi-spine (3 stars, last pushed yesterday), licensed MIT. It adds 14 tokens to every session and 1,616 once invoked, about $0.0001 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 2026-08-31.
reviewer compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| reviewerbeettlle/pi-spine | Agent | 3 | yesterday | MIT | A | 14 | Claude Code |
| cpp-revieweraffaan-m/ECC | Agent | 299k | yesterday | MIT | A | 41 | Claude Code, Kiro |
| thermo-nuclear-code-quality-review-subagentmichael-denyer/pstack-claude | Agent | 4.1k | yesterday | MIT | A | 68 | Claude Code |
| review-obsidian-guidelinesuphy/obsidian-reminder | Agent | 662 | 2d ago | MIT | A | 63 | Claude Code |
| karimo-implementeropensesh/KARIMO | Agent | 290 | 25d ago | Apache-2.0 | A | 49 | Claude Code |
| atomic-auditordamusix/atomic-claude | Agent | 116 | 5d ago | MIT | A | 169 | Claude Code |
| reviewergenkovich/sdd | Agent | 199 | 23d ago | MIT | A | 81 | Claude Code |