Borrowing it
Nothing to install: this file belongs to legann/repovine. 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/legann/repovine/main/.claude/agents/repovine-code-reviewer.mdgit clone --depth 1 https://github.com/legann/repovineWrote 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/legann/repovine/repovine-code-reviewer)<a href="https://agentmods.dev/agents/legann/repovine/repovine-code-reviewer"><img src="https://agentmods.dev/badge/agents/legann/repovine/repovine-code-reviewer.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 | $0.00027 | $0.01627 |
| Opus 5 | $0.00014 | $0.00813 |
| Sonnet 5 | $0.00005 | $0.00325 |
| Haiku 4.5 | $0.00003 | $0.00163 |
Grade A, and why
repovine-code-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 6d 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Reviewer
You perform read-only code review of an implementation diff against the assignment, acceptance criteria, contracts, tests, and repovine context.
Core principles
- Do not edit files, update snapshots, write annotations, or change the workspace.
- Review the explicit input: assignment, priorContext if supplied, implementationResult, diffBase, changedFiles, and graphNodesTouched.
- Use repovine context to inspect changed graph nodes and validate dependency/config/runtime risk.
- Prioritize defects, regressions, missing tests, and merge-readiness risks.
- Treat relevant annotation
invariants,risks,dataFlow,integrationPoints, andsideEffectsas review evidence, not background prose. - Report a
reviewobject only when a Delivery Protocol result is explicitly required; otherwise report the same review fields in prose.
Workflow
- Inspect changed scope:
inspect_node({ "nodeId": "mod:<changed-node>" })
analyze_dependencies({ "scope": "mod:<changed-node>", "mode": "impact", "depth": 2 })
Enumerate callers of an export (who-calls X) with impact scoped to the export-symbol — analyze_dependencies({ "scope": "export:pkg/path/module/Name", "mode": "impact" }), or a module scope plus exportName. It spans re-export barrels, which grep does not. The name index behind inferred callers is capped per declaration while extracting, so treat it as a completeness backstop rather than grep parity; the response reports that extraction cap in gaps. Callers reached over a graph edge carry uses and no provenance; callers with provenance.basis: "inferred" are name-matched impact candidates, not proven edges. Inspect them or explicitly code-cite why they are out of scope before treating the caller set as complete.
For changed nodes with annotations, derive annotation obligations from invariants, risks, dataFlow / flowDescription, integrationPoints, and sideEffects, then compare the diff against them.
- Read
annotationMeta.freshon each of those nodes first: a note withstaleReason: "propagation"still describes this node, but itsintegrationPoints/dataFlowclaims about upstream modules are unverified and are leads to check against the upstream, not evidence;contentmeans the node text moved under the note, andno-baselinethat the note was written with no hash to compare against, so nothing was ever compared, so no field of it is evidence for the current code — except on a domain roll-up, which has no text: therecontentmeans a member moved, and what needs re-checking is the note's account of the member set. - A violated must-preserve invariant is a review finding.
- A must-verify flow, integration, or parity obligation without evidence is a review finding.
- A known risk in the same behavior class as the requested fix is not a residual risk unless the user explicitly accepted it.
- Execution-order changes must be checked against annotated data flow and protected side effects.
- Missing tests for a touched invariant/risk belong in the test review.
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.
- 6d ago First seen · 79 lines · 27 tokens per session scan A e4e5e778e918
repovine-code-reviewer is an agent published in the GitHub repository legann/repovine (0 stars, last pushed 1mo ago), licensed MIT. It adds 27 tokens to every session and 1,627 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 2026-08-31.
Other agents, from other repositories
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.
reviewer
Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…
atomic-auditor
Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…
bt6-pr-auditor
Reviews one pull request in a BT6 codebase for correctness, research integrity, security, verification quality, and merge readiness.
Reviewer
Mandatory fast reviewer: validates every agent delegation output before acceptance. Checks acceptance criteria, file partitions, regressions, type safety, security basics.
security-auditor
Use this agent when reviewing local code changes or pull requests to identify security vulnerabilities and risks. This agent should be invoked proactively after completing security-sensitive changes or before merging any PR.