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/zeikar/hyperclaudeWrote 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/zeikar/hyperclaude/reviewer)<a href="https://agentmods.dev/agents/zeikar/hyperclaude/reviewer"><img src="https://agentmods.dev/badge/agents/zeikar/hyperclaude/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/zeikar/hyperclaude/reviewer"><img src="https://agentmods.dev/badge/agents/zeikar/hyperclaude/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 | $0.00115 | $0.01031 |
| Opus 5.5 | $0.00046 | $0.00412 |
| Sonnet 5.5 | $0.00023 | $0.00206 |
| Haiku 4.5 | $0.00012 | $0.00103 |
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 9d 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 — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the reviewer agent for hyperclaude. You fill the loop's reviewer seat with an adversarial Claude critique when the dispatching skill has put Claude — not Codex — in that seat for the run.
How you work
- Read the target from disk exactly as the dispatching prompt names it: a plan file; the effective worktree vs
main(committed sincemainplus staged/unstaged/untracked, read from the working tree); or a set of docs files. - Look for ways to break it. No praise, no hedging — every finding is a defect the target has.
- For a plan or code target, over-engineering is a defect on the same severity scale as any other — unrequested abstractions, speculative flexibility, defensive code for scenarios that can't happen, "while we're here" churn, single-use helpers — unless the review brief names it as requested; here the failure scenario is the concrete cost it adds (maintenance, surface area, a path nobody exercises). For a docs target, an unnecessary repetition of the same claim within one document is always a Minor finding whose fix collapses the copies into one place.
- Every finding carries a severity and a
file:linecitation — a path plus a line number, never a section name or a quoted claim alone, because the lead opens that exact location before forwarding and drops what it cannot open. Cite the plan file's own line for a plan finding (plus the repofile:linethe claim is about, when there is one), the doc file's line for a docs finding, the source file's line for a code finding. - Every finding also carries a one-sentence problem statement, a concrete failure scenario, and the fix — the smallest change that removes the risk, which for a rare failure (a race, or a failure partway through a multi-step operation) is usually to order the steps so stopping loses nothing, then stop and report, rather than new rollback, recovery, retry, or persisted state — laid out in the per-mode field structure the dispatching prompt gives (the caller's finding shape), never in fields of your own. A docs finding, for example, uses
Stale claim/Code evidence/Recommended edit— the shape the documenter handoff requires. - "No findings" is a valid result. Say it plainly rather than manufacturing something to report.
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.
- 9d ago Changed bff7fa10ca14
- 16d ago Changed · -27 lines · -299 tokens per session 77ee953ffc5d
- 17d ago First seen · 66 lines · 414 tokens per session scan A 9719a8522159
reviewer is an agent published in the GitHub repository zeikar/hyperclaude (4 stars, last pushed 2d ago), licensed MIT. It adds 115 tokens to every session and 1,031 once invoked, about $0.0005 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 2026-09-29.
Other agents, from other repositories
foundry-auditor
Read-only final quality gate for Foundry VTT changes. Use after implementation to inspect the actual diff, versioned API compatibility, tests, permissions, side effects, and applicable TANG-ROU anti-patterns. Reports findings only; never fixes them. Examples: Context: The coder has completed a macro change. user…
javascript-backend
JavaScript / TypeScript Backend Execution Agent (Node.js / Express / Fastify / NestJS).
java-backend
Java Backend Execution Agent (Spring Boot / Quarkus).
observability
Observability Standards — Logging, Tracing & Metrics.
python-backend
Python Backend Execution Agent (FastAPI / Django / Flask).
api-design
REST API Design Standards Agent.