Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/ReviewToolkits/cext-review-toolkitnpx agentmods add commands/reviewtoolkits/cext-review-toolkit/exploreWrote 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/commands/reviewtoolkits/cext-review-toolkit/explore)<a href="https://agentmods.dev/commands/reviewtoolkits/cext-review-toolkit/explore"><img src="https://agentmods.dev/badge/commands/reviewtoolkits/cext-review-toolkit/explore/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/commands/reviewtoolkits/cext-review-toolkit/explore"><img src="https://agentmods.dev/badge/commands/reviewtoolkits/cext-review-toolkit/explore.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 | $0.00087 | $0.04024 |
| Opus 5.5 | $0.00035 | $0.01610 |
| Sonnet 5.5 | $0.00017 | $0.00805 |
| Haiku 4.5 | $0.00009 | $0.00402 |
Grade A, and why
explore scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 337 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Comprehensive C Extension Analysis
Run a comprehensive analysis of a CPython C extension using multiple specialized agents, each focusing on a different aspect of extension correctness. Extension discovery runs first to understand the project layout.
Arguments: "$ARGUMENTS"
Plugin root: <plugin_root> refers to the directory containing this command file's parent -- i.e., the plugins/cext-review-toolkit/ directory. Resolve it relative to this file's location.
Argument Parsing
Parse arguments into three categories:
Scope (path or glob):
.or omitted → entire project (default)src/→ specific directory treesrc/myext.c→ specific file
Aspects (which agents to run):
refcounts→ refcount-auditorerrors→ error-path-analyzernull-safety→ null-safety-scannergil→ gil-discipline-checkermodule-state→ module-state-checkertype-slots→ type-slot-checkerabi→ stable-abi-checkercompat→ version-compat-scannerpyerr-clear→ pyerr-clear-auditorresources→ resource-lifecycle-checkerparity→ parity-checkercomplexity→ c-complexity-analyzerhistory→ git-history-analyzerexternal-tools→ run external tools only (clang-tidy, cppcheck)all→ all agents (default)
Options:
deep→ full detail, no output truncationsummary→ summary tier only (faster)parallel→ run agents concurrently where possible--max-parallel N→ cap concurrent agents per group (default: 2)--runs N→ run agents N times (default: 1). Run 2+ are independent naive passes. Findings are deduplicated across runs. If run 2 finds 0 new findings, report "high confidence — converged on first pass."--informed-reruns→ when used with--runs 3, run 3 agents receive a summary of runs 1-2 findings with the instruction: "These bugs were already found. Look at ADJACENT code, unexplored error paths, and patterns similar to these confirmed bugs that the prior runs might have missed." This targets the informed-rerun technique that found 33% more bugs on simplejson.
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 · 337 lines · 87 tokens per session scan A a41a7efd6a36
explore is a command published in the GitHub repository ReviewToolkits/cext-review-toolkit (29 stars, last pushed 2mo ago), licensed MIT. It adds 87 tokens to every session and 4,024 once invoked, about $0.0003 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 2026-08-30.
Other commands, from other repositories
cpp-build
Fix C++ build errors, CMake issues, and linker problems incrementally. Invokes the cpp-build-resolver agent for minimal, surgical fixes.
crash-analysis
Autonomous crash root-cause analysis for C/C++ bugs.
gf-fix
Automatically fix lint errors in a SystemVerilog file using iterative lint-fix cycles.
cpp-build
Fix C++ build errors, CMake issues, and linker problems incrementally. Invokes the cpp-build-resolver agent for minimal, surgical fixes.
tron-debug
We are debugging the C++ produced by the ingest pipeline from Torch Fx. You can read the Note in @src/Fx.hs for details on the Torch -> Bulk pipeline. After that it passes through Bulk, Loopy, Tron, and CPP intermediate representations.
cpp-build
Fix C++ build errors, CMake issues, and linker problems incrementally. Invokes the cpp-build-resolver agent for minimal, surgical fixes.