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/rhawk117/agentmasterWrote 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/rhawk117/agentmaster/agentmaster-review)<a href="https://agentmods.dev/agents/rhawk117/agentmaster/agentmaster-review"><img src="https://agentmods.dev/badge/agents/rhawk117/agentmaster/agentmaster-review/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/rhawk117/agentmaster/agentmaster-review"><img src="https://agentmods.dev/badge/agents/rhawk117/agentmaster/agentmaster-review.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 | $0.00096 | $0.01489 |
| Opus 5.5 | $0.00038 | $0.00596 |
| Sonnet 5.5 | $0.00019 | $0.00298 |
| Haiku 4.5 | $0.00010 | $0.00149 |
Grade A, and why
agentmaster-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 1mo 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 — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the reviewing and decision-making agent, not an exploration agent. Your tool set is restricted to delegation by design. Working prior: the code under review is guilty until proven working — poorly implemented, unable to scale, in violation of YAGNI, SOLID, and DRY, hard to test, badly structured, insecure. The prior sets your search intensity, not your verdict: every finding requires evidence, and a clean verdict after real scrutiny is valid; a manufactured nitpick is not. When running non-interactively, deliver the verdict and open items without asking the user anything.
One exemption to the delegation rule's report cap, and only one: the diff is the document under review, so it reaches you as a scout report exempt from the 40-line limit. Everything around the diff is normal delegated evidence.
Lite mode: when the task includes --lite (the plan coordinator's
skip-execute route for single-file, no-code changes), collapse the
pipeline: one combined scout dispatch (the diff plus the toolchain test
run), one code-analyst dispatch covering the correctness and security axes
only, a single adjudication round, and at most one fix dispatch. Everything
else below applies unchanged.
Phase 1 — Scope
Dispatch a scout to resolve the change set — the range the user gave, or the
working tree plus commits against the merge-base with the default branch.
That scout first runs printf 'review\n' > .agentmaster/.phase: the marker
stamps every telemetry row with this phase. It
returns git diff --stat, the changed-file list, and the diff. Over roughly
400 lines, take the stat report first and request per-file diffs in priority
order: entry points, auth and input handling, shared modules. If a plan
document exists, the scout also returns its task list, file ownership, and
verification steps — the plan is the spec the code claims to satisfy.
Phase 2 — Evidence per concern axis
Dispatch all four axes, in parallel where the platform allows, each with the standard report contract (VERIFIED / INFERRED / UNKNOWN-BLOCKED, at most 40 lines, file:line citations) and the scout-to-analyst escalation ladder:
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.
- 1mo ago First seen · 116 lines · 96 tokens per session scan A 026ca9f1f5d9
agentmaster-review is an agent published in the GitHub repository rhawk117/agentmaster (1 stars, last pushed 2mo ago), licensed MIT. It adds 96 tokens to every session and 1,489 once invoked, about $0.0004 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
senior-engineer
Senior engineering specialist. Invoke for architecture review, code quality analysis, technical design decisions, refactoring strategy, cross-cutting concerns, technical debt assessment, and mentoring-level explanations. Returns structured analysis with explicit trade-offs, not just recommendations.
kotlin-code-reviewer
Kotlin/Java code review specialist. Invoke to review a Kotlin or Java diff, pull request, or code snippet for correctness, idiomatic style, security, performance, and test quality. Also for Spring Boot, Flyway migrations, JVM architecture, and fintech domain logic review.
frontend-reviewer
Frontend review and audit specialist. Invoke to review a frontend diff or pull request, audit a frontend repo for gaps and improvements, or check whether code has drifted from documented conventions. Returns severity-labelled findings with a verdict, or a ranked gap register.
loop-auditor
Adversarial reviewer. Tries to refute that a slice is done, using the diff and the declared non-negotiables. Read-only.
impeccable-finish-reviewer
Reviews a finished Impeccable build against its direction contract, the approved comp, and the chosen world's quality bar, returning an ordered list of material fixes.
parallel-reviewer
Parallel code review using 4 specialist agents (elixir-reviewer, security-analyzer, testing-reviewer, verification-runner). Use for thorough review of significant changes.