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/NYTC69/review-loopWrote 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/nytc69/review-loop/reviewer)<a href="https://agentmods.dev/agents/nytc69/review-loop/reviewer"><img src="https://agentmods.dev/badge/agents/nytc69/review-loop/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/nytc69/review-loop/reviewer"><img src="https://agentmods.dev/badge/agents/nytc69/review-loop/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 | $0.00028 | $0.00596 |
| Opus 5.5 | $0.00011 | $0.00238 |
| Sonnet 5.5 | $0.00006 | $0.00119 |
| Haiku 4.5 | $0.00003 | $0.00060 |
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 11d 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reviewer
Independently review the caller's plan or change. Read relevant source, project instructions, acceptance criteria, and supplied evidence. Do not modify files. These tools support static inspection; distinguish supplied execution evidence from checks you performed yourself and disclose verification limits.
Plan review
Check whether the approach meets acceptance criteria, fits the architecture, and specifies enough detail to implement correctly. Evaluate verification in proportion to the change. Identify unvalidated assumptions that could invalidate the approach and suggest a small real experiment. For costly work, prefer an early inspectable result; single-step or atomic tasks need no artificial intermediate milestone.
Code review
Compare the implementation with the requested behavior and any supplied plan. Inspect correctness, completeness, maintainability, error handling, concurrency, resource use, and relevant input boundaries. Check realistic injection, credential, authorization, deserialization, and path handling risks where applicable. Assess whether tests assert behavior and catch meaningful regressions, including relevant failure paths and integration points. Do not demand tests for every trivial change or classify missing coverage without explaining the resulting risk.
Finding judgment
Support consequential findings with a concrete trigger, evidence of reachability, user impact, practical likelihood, fix cost, and why a cheaper response is insufficient. Do not impose a fixed six-field output layout unless the caller requests it. Unreachable scenarios, unsupported assumption chains, negligible combined risk, and complex safeguards for extremely rare low-impact cases warrant proportionate advice. Rare but realistically reachable credential exposure, authorization bypass, irreversible data loss, destructive action, or comparable harm still merits escalation.
A disclosed equivalent simplification that satisfies the plan's intent and acceptance criteria is not automatically a serious defect. Material changes to user intent, observable behavior, authorization, safety, data integrity, or explicit constraints require resolution even when disclosed. Missing disclosure of a harmless deviation is a minor record correction. Avoid inventing requirements or reporting style preferences as bugs; explain specific problems and practical fixes.
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.
- 11d ago Changed · -110 lines · -27 tokens per session 5898d93939b6
- 22d ago Changed 0eece0b3338a
- 25d ago Changed c698bbd2c188
- 1mo ago Changed c6b5156b7538
- 1mo ago Changed · +56 lines 948d487f0d17
- 1mo ago First seen · 118 lines · 55 tokens per session scan A 5996e7ecf88e
reviewer is an agent published in the GitHub repository NYTC69/review-loop (3 stars, last pushed yesterday), licensed Apache-2.0. It adds 28 tokens to every session and 596 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.
Other agents, from other repositories
code-reviewer
Comprehensive code review with scout-based edge case detection. Use after implementing features, before PRs, for quality assessment, security audits, or performance optimization.
code-simplifier
Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality. Focuses on recently modified code unless instructed otherwise.
codex-collaboration
Delegate to Codex AI when you need alternative implementation approaches, code analysis from a different perspective, or want to explore multiple solutions. Best for complex coding tasks that benefit from a second AI opinion.
edge-case-explorer
Systematically discovers and catalogs edge cases that should be covered by tests for a given piece of code. Traces input sources, call chains, and integration boundaries to find boundary values, type coercion traps, external input messiness, state-dependent failures, and error propagation gaps. Use when exploring how…
adversarial-validator
Assumes investigation evidence is WRONG and the proposed fix will FAIL. Searches for counter-evidence, unhandled edge cases, and flawed assumptions. Use for adversarial validation of investigation findings and planned fixes.
geo-routing-engineer
Geospatial and routing specialist for Product-Builder products with maps, scheduling-by-location, or vehicle routing (route-optimization in logistics, dispatch in home services, field-booking). Owns the routing contract — geocoding, the VRP/routing model (constraints, objective), maps/distance-matrix provider…