roborev is a background code-review system for AI coding agents that checks each Git commit and records findings in a review database. It helps developers catch issues in generated code while they work. Catalogue add-ons connect its findings to agent workflows and can support iterative review and fixing.
Borrowing it
Nothing to install: this file belongs to kenn-io/roborev. 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/kenn-io/roborev/main/AGENTS.mdgit clone --depth 1 https://github.com/kenn-io/roborevWrote 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/instructions/kenn-io/roborev/agents-md)<a href="https://agentmods.dev/instructions/kenn-io/roborev/agents-md"><img src="https://agentmods.dev/badge/instructions/kenn-io/roborev/agents-md/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/instructions/kenn-io/roborev/agents-md"><img src="https://agentmods.dev/badge/instructions/kenn-io/roborev/agents-md.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 | $0.05631 | $0.05631 |
| Opus 5.5 | $0.02252 | $0.02252 |
| Sonnet 5.5 | $0.01126 | $0.01126 |
| Haiku 4.5 | $0.00563 | $0.00563 |
Grade B, and why
roborev AGENTS.md scanned grade B with 1 finding 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 7d 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.
Unrestricted tool accessmediumExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
- Release workflow testing must not publish anything. Do not run any command, workflow mode, or release tool path that creates public How it starts
The opening of the file, as written. The whole thing — 346 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Purpose
This repo hosts roborev, a local daemon + CLI for AI-assisted code review. Use this guide to navigate the codebase quickly, understand subsystem boundaries, and avoid missing adjacent changes when working in a growing Go project.
Start Here
- Product behavior and user-facing workflows:
README.md - CLI entry point and command registration:
cmd/roborev/main.go - Command discovery:
rg '^func .*Cmd\(' cmd/roborev - Daemon HTTP API and route wiring:
internal/daemon/server.go - Worker pool and job execution:
internal/daemon/worker.go - SQLite schema migrations:
internal/storage/migrations.go - Config loading and resolution:
internal/config/config.go - Prompt construction:
internal/prompt/prompt.go - TUI entry point:
cmd/roborev/tui/tui.go - Release documentation workflow:
docs/README.md
Documentation Style
- Write documentation for humans and agents in plain language: name who does what, use short sentences, and explain unfamiliar terms. Prefer bullets for independent facts, numbered steps for sequences, and paragraphs for rationale.
- Living architecture describes the system, not the history of implementing it. Separate implemented behavior, approved-but-unbuilt work, proposals, and historical decisions. Do not turn every document into a mandatory template.
- Organize around reader questions: purpose, what works today, responsibilities, how it works, rules and failure behavior, and open decisions. Use only the sections the topic needs. Keep detailed protocol definitions below the overview or in focused linked documents.
- Make the opening useful on its own. State the purpose briefly and list current capabilities and gaps separately; do not pack status, ownership, restrictions, and future work into one paragraph.
- Give each bullet one main idea. Use a small table for repeated comparisons and a diagram when it clarifies a relationship or sequence. Neither is required decoration, and a long paragraph does not become readable by adding a bullet.
- State rules directly, then explain non-obvious reasons. Preserve exact field names, authorization checks, failure conditions, and protocol requirements; simpler wording must not weaken the contract.
- Update the relevant section instead of appending a narration of the latest change. Give each fact one owning document and link to it elsewhere. Indexes should route readers, not duplicate detailed implementation status.
- Keep superseded designs outside the normal reading path in clearly marked historical sections or linked documents. Preserve decision rationale, approvals, and active exceptions, including their removal conditions.
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.
- 7d ago Changed · +2 lines · +31 tokens per session f0d61d487509
- 11d ago Changed · +1 lines · +107 tokens per session 71ccb468567c
- 16d ago Changed · +49 tokens per session c27ceb45fa2f
- 19d ago Changed · +4 lines · +50 tokens per session dbc130bc4959
- 21d ago Changed · +5 lines · +51 tokens per session 46d1205b0867
- 29d ago Changed · +1 lines · +53 tokens per session ebceb437f350
- 1mo ago Changed · +33 lines · +427 tokens per session ee75633f0d3b
- 1mo ago Changed · +1 lines · +45 tokens per session 12a7c5b1f1da
- 1mo ago Changed · +4 lines · +61 tokens per session a01544b7ad2b
- 1mo ago First seen · 295 lines · 4,757 tokens per session scan B 44a6df6cb595
roborev AGENTS.md is an instructions file published in the GitHub repository kenn-io/roborev (1,771 stars, last pushed today), licensed MIT. It adds 5,631 tokens to every session, about $0.0225 per session on Opus 5.5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.