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/Stickman230/claude-pentestWrote 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/stickman230/claude-pentest/pentest-exit)<a href="https://agentmods.dev/commands/stickman230/claude-pentest/pentest-exit"><img src="https://agentmods.dev/badge/commands/stickman230/claude-pentest/pentest-exit/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/stickman230/claude-pentest/pentest-exit"><img src="https://agentmods.dev/badge/commands/stickman230/claude-pentest/pentest-exit.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 | $0.00026 | $0.00945 |
| Opus 5 | $0.00013 | $0.00473 |
| Sonnet 5 | $0.00005 | $0.00189 |
| Haiku 4.5 | $0.00003 | $0.00094 |
Grade A, and why
pentest-exit 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 — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Output the following closing banner verbatim:
============================================================
[PENTEST SESSION CLOSING]
============================================================
Step 1 — Identify Engagement
Use AskUserQuestion to ask:
"What is the engagement name for this session? (the name used in outputs/{name}/ — check your engagement summary or activity logs)"
Store the answer as {engagement_name}.
Step 2 — Findings Review
You MUST complete this step before outputting the session summary. Do not skip it even if the engagement folder appears empty.
First, try Schema A paths:
Read all files matching: outputs/{engagement_name}/findings/finding-*/description.md
Also read outputs/{engagement_name}/pentest-report.json if it exists.
If outputs/{engagement_name}/findings/ is empty or does not exist, try Schema B paths:
Read all files matching: outputs/{engagement_name}/processed/findings/finding-*/description.md
Also read outputs/{engagement_name}/report/pentest-report.json if it exists.
Use the schema that produced findings. If both are empty, record: "No findings documented in outputs/{engagement_name}/"
From the files found:
- Count findings by severity: Critical, High, Medium, Low, Info
- Note the title or one-line description of each finding
Step 3 — Notes Flush
You MUST complete this step before outputting the session summary. Do not skip it.
Check whether any findings or notes are held in the current conversation context but NOT yet written to disk. This includes:
- Any finding analysis discussed but not saved
- Any vulnerability details drafted in conversation but without a corresponding description.md
- Any activity summaries not yet appended to logs
If any such in-progress work exists, write it to disk now:
- Findings go to:
outputs/{engagement_name}/findings/finding-{NNN}/description.md(where NNN is the next sequential number) - Activity logs go to:
outputs/{engagement_name}/activity/ - Final report (if complete) goes to:
outputs/{engagement_name}/pentest-report.json
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 First seen · 110 lines · 26 tokens per session scan A 7e7675eb997b
pentest-exit is a command published in the GitHub repository Stickman230/claude-pentest (100 stars, last pushed 3mo ago), licensed MIT. It adds 26 tokens to every session and 945 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 2026-08-30.
Other commands, from other repositories
unpack
Detect and remove binary packing/protection.
dashboard
Show running scans, system health, and engagement status.
apiscan
API security audit — REST, GraphQL, JWT analysis, parameter discovery.
payload
AI-powered payload generation for XSS, SQLi, LFI, command injection, and more.
recon
Full reconnaissance workflow — subdomain enum, tech detection, WAF fingerprinting, crawling.
bruteforce
Password brute force and hash cracking against target services.