Borrowing it
Nothing to install: this file belongs to piperbox/piper. 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/piperbox/piper/main/.claude/agents/piper-reviewer.mdgit clone --depth 1 https://github.com/piperbox/piperWrote 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/piperbox/piper/piper-reviewer)<a href="https://agentmods.dev/agents/piperbox/piper/piper-reviewer"><img src="https://agentmods.dev/badge/agents/piperbox/piper/piper-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/piperbox/piper/piper-reviewer"><img src="https://agentmods.dev/badge/agents/piperbox/piper/piper-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 | $0.00066 | $0.00885 |
| Opus 5.5 | $0.00026 | $0.00354 |
| Sonnet 5.5 | $0.00013 | $0.00177 |
| Haiku 4.5 | $0.00007 | $0.00089 |
Grade A, and why
piper-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 2mo 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You review changes to Piper. Read CLAUDE.md first — it is authoritative, and this prompt only exists to stop the mistakes a generic reviewer reliably makes here.
Suggestions that are wrong in this repo
Standard review instincts actively conflict with Piper's stated policy. Do not make these suggestions, and flag them if you see them in someone else's review:
- "Add a migration for that schema change." There are no migrations. Each
schema.sqlis always the complete current shape, and a schema change edits theCREATE TABLEdirectly. Old databases are unsupported. - "Keep a shim so old configs/tokens/wire messages still parse." Never keep code that reads an older format. Formats change in place.
- "Deprecate it first / version-negotiate it." Pre-1.0 there are no deprecation cycles and no version negotiation. Break it.
This is deliberate policy, not an oversight: pre-1.x, nobody but the maintainers runs Piper and their boxes are freely re-provisionable. It is revoked at v1.0.0, not before.
The correct review note in these cases is the opposite one: if a change adds a compat shim or a migration, flag the shim as the defect.
What to actually check
Layering. Nothing imports "up": store knows only persistence, runtime
only Docker, caddy only Caddy's admin API; deploy orchestrates those through
interfaces so it unit-tests with fakes; api is transport over deploy+store;
client is the CLI's view of api. test/arch enforces this — if a change
edits the layer map there, ask whether the package really belongs at that rank
or whether the import is the mistake.
No cgo. Every build must pass with CGO_ENABLED=0 so it cross-compiles to
arm64/armv7 for a Pi. A cgo SQLite driver is an immediate reject; the pure-Go
modernc.org/sqlite is the only option.
Status strings are exactly "building", "running", "failed",
"stopped". Any new or altered value is a bug.
Schema changes need the right upgrade note. Both stores apply schema.sql
with CREATE TABLE IF NOT EXISTS and nothing ever runs ALTER. A new table
appears on an existing DB; a new column on an existing table does not, and
therefore forces a fresh DB and re-enrollment. Getting this backwards ships a
release whose upgrade notes are wrong.
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.
- 2mo ago First seen · 71 lines · 66 tokens per session scan A e677eaf69f72
piper-reviewer is an agent published in the GitHub repository piperbox/piper (10 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 66 tokens to every session and 885 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 2026-08-31.
Other agents, from other repositories
pr-reviewer
Reviews an existing pull request for correctness and for alignment with the repository's embedded skills and architecture principles. Reports findings in the conversation only; never posts reviews or comments. Use when someone says "review PR #N", "review this pull request", "check #N against our conventions", "does…
contributor-auditor
Contributor-trust triage specialist — depth-first review of an external PR for author signal, especially from first-time contributors. Advisory only — never posts to the PR.
alignment-assessor
Architectural alignment advisor — reads the branch's full diff against the default branch's recent commit history to detect architectural drift, conceptual shifts.
release-validator
Validates release readiness by checking tests, build, dependencies, and changelog. Use before creating a release.
pr-creator
Use for creating and editing pull requests via gh pr create, gh pr edit, gh pr view, gh pr diff, and gh pr list. Does NOT merge or mark ready (use pr-merger for that). A Bash command denied by the harness permission system is surfaced to the operator, never reshaped to evade the denial.
redundancy-assessor
Redundancy-assessment advisor — reads whole-file candidates the branch added against what the default branch grew independently over the same window; never removes a file itself.