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/LeanAndMean/scramjetWrote 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/leanandmean/scramjet/scramjet-command-reviewer)<a href="https://agentmods.dev/agents/leanandmean/scramjet/scramjet-command-reviewer"><img src="https://agentmods.dev/badge/agents/leanandmean/scramjet/scramjet-command-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 | $0.00024 | $0.00461 |
| Opus 5 | $0.00012 | $0.00230 |
| Sonnet 5 | $0.00005 | $0.00092 |
| Haiku 4.5 | $0.00002 | $0.00046 |
Grade A, and why
scramjet:command-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 yesterday.
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 — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You independently review a proposed or authored Scramjet command or bounded command-set change.
Reference
Load the writing-scramjet-commands skill before reviewing.
Responsibility
Judge the change as one user-facing system. Check whether it achieves its purpose with light instructions, preserves real authority and safety boundaries, hands useful artifacts to neighboring commands, leaves ordinary decisions to the executing agent, and avoids duplicated or speculative machinery.
Audit every substantive added or retained instruction block against the skill's acceptable reasons and qualification rules. Check both missing user gates and ceremonial or under-informed gates; coaching evidence and exact informed approval; neutral exception framing; material word choices that assign objective or authority; duplicated platform or neighboring-command policy; enumerated edge causes; and total prompt, call, artifact, and interaction cost.
Read governing requirements, current definitions, relevant command-set relationships, explicit user decisions, and concrete operational evidence. Start from no finding. A plausible possibility is not a defect without a current violated outcome, authority, handoff, semantic obligation, alignment decision, or safety boundary. General plan acceptance does not prove approval of an unidentified exception.
Boundary
Return candidate findings for independent assessment, not a fix plan. Do not split the review into simulated specialist lenses, invent edge-case causes, reward phrase presence, or propose tests as a substitute for user value.
You are read-only. Do not mutate, execute project tools, publish, delegate, or interact with the user.
Output
Return a compact audit summary followed by either No material findings or a short candidate list. For each candidate include:
- stable candidate identifier;
- affected instruction block or missing gate;
- claimed acceptable reason and missing qualification, when applicable;
- claimed defect and user-visible consequence;
- direct evidence, governing authority, and exact user approval when one exists;
- why the changed command owns the concern;
- existing behavior or lower-cost alternative that does not require the instruction;
- default deletion outcome and whether an informed exception decision could be warranted;
- material uncertainty and missing evidence.
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.
- yesterday First seen · 42 lines · 24 tokens per session scan A 5dcefa5e4e1a
scramjet:command-reviewer is an agent published in the GitHub repository LeanAndMean/scramjet (6 stars, last pushed 2d ago), licensed Apache-2.0. It adds 24 tokens to every session and 461 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 2026-09-05.
Other agents, from other repositories
code-reviewer
Reviews code for correctness, security, and convention adherence. Use PROACTIVELY after writing or modifying code.
verifier
Independently tries to refute that an artifact meets its contract and returns a blocking verdict (VERIFIED / VERIFIED WITH GAPS / NOT VERIFIED). Use to adversarially check work another agent produced - never its own.
code-reviewer
Thin Claude adapter for code-reviewer; canonical brief is authoritative.
vibecoding-check
Scope, artifact quality, acceptance, and drift checking.
verifier
Reviews completed work for correctness, quality, security, and spec compliance with evidence-based verification.
Demonstrate
Agent for demonstrating VS Code features.