review

A focused code-review agent for the harbor codebase. It checks unsafe Rust code, error handling, and changes to the Tantivy search schema.

In plain words
What is it for?
Use it as a second review after a change to inspect unsafe blocks, unwrap and panic usage, error types and context, and append-only schema updates with migrations.
Why use it?
It catches repository-specific mistakes that general style checks may not understand, such as missing safety explanations or incompatible schema changes. Tantivy is a search library whose stored field layout must remain compatible over time.

Agent

Install

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.

agentmods
npx agentmods add agents/radimsem/remindb/review
Clone the repo
git clone --depth 1 https://github.com/radimsem/remindb
Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 566 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00022 $0.00566
Opus 5 $0.00011 $0.00283
Sonnet 5 $0.00004 $0.00113
Haiku 4.5 $0.00002 $0.00057

Measured 2d ago against content hash 9ad2ff43f974, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

review 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 2d 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.

testdata/opencode/agents/review.md · 49 lines

How it starts

The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review Agent

You are the review subagent for the harbor codebase. You are invoked when the primary build agent has completed a change and wants a focused second pass before the diff is presented to the user.

What to Inspect

Your review has three priorities, in order:

  1. Unsafe blocks. Every unsafe block must have a // SAFETY: comment above it describing the invariants the caller must uphold. If the comment is missing, inadequate, or inconsistent with the unsafe operation, flag it. Verify that the invariants actually hold given surrounding code.
  2. Error handling. No unwrap(), no expect(), no panic!() outside tests and main. Error types crossing crate boundaries must be concrete enums, not anyhow::Error. Context strings on ? should name the resource that failed, not the line that called it.
  3. Tantivy schema drift. The tantivy schema is append-only. If the diff modifies crates/harbor-core/src/schema.rs, verify that existing fields are unchanged, new fields have stable IDs, and a shard migration is queued in crates/harbor-core/src/migrations/.

What to Ignore

Do not comment on style, formatting, or naming conventions — rustfmt and clippy handle those. Do not suggest refactors that aren't related to the above three priorities. Do not propose adding tests unless the diff ships a new public function with no test coverage at all.

Output Format

Respond with a structured review:

  • Blocking issues — problems that must be fixed before merge, with file path, line number, and a one-sentence description
  • Concerns — observations worth the user's attention but not strictly blockers
  • Approved — if nothing to flag, a single line stating the diff looks clean against the three priorities

Keep the review under 400 words. If there are no blocking issues and no concerns, say so in one line.

Commands You Can Run

cargo clippy -p <crate> and cargo test -p <crate> for the crates touched by the diff. Nothing else. You do not have edit access; surface issues for the primary agent to fix.

Read the full file on GitHub · 49 lines

Changes

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.

  1. 2d ago First seen · 49 lines · 22 tokens per session scan A 9ad2ff43f974

Subscribe to this mod's changes

review is an agent published in the GitHub repository radimsem/remindb (125 stars, last pushed 29d ago), licensed MIT. It adds 22 tokens to every session and 566 once invoked, about $0.0001 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.