agentleFS
jaskaranhundaljaskaranhundal/usap-skills/pentest/CLAUDE.md

usap-skills / pentest

CLAUDE.md4 starsChanged 2 months agoApache-2.0View on GitHub

The Pentest domain contains skills for conducting authorized penetration testing and compiling the results into stakeholder-ready reports. Skills span structured web application assessment (OWASP Testing Guide v4.2, OWASP Top 10 2021) and penetration test report generation (PTES, OWASP reporting standards). Pentest skills operate strictly within written authorization and defined Rules of Engagement. They classify findings by OWASP category, CVSS v3.1, and MITRE ATT&CK, assess real-world exploitability, and produce sanitized remediation guidance — never weaponized exploit code. Active-testing skills set human_approval_required:…

The file

We split this CLAUDE.md into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    All skill paths are relative from the repository root as pentest/<slug>/.

    line 22
  2. Distinctive:

    No active testing without confirmed written authorization (target, test window, Rules of Engagement).

    line 37
  3. Distinctive:

    No weaponized exploit code in any output — proof-of-concept material is sanitized to reproduction steps.

    line 39
  4. Very specific:

    Validate scope and authorization as step zero: No reconnaissance, scanning, or exploitation guidance is produced until written …

    line 47
  5. Distinctive:

    Deduplicate across sources before reporting: When findings arrive from multiple skills (web-app-pentest, network-pentest, …

    line 51

≈ “Common” is an estimate: at least 1 in 5 of the 27,641 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

Pentest Domain — CLAUDE.md

Purpose

The Pentest domain contains skills for conducting authorized penetration testing and compiling the results into stakeholder-ready reports. Skills span structured web application assessment (OWASP Testing Guide v4.2, OWASP Top 10 2021) and penetration test report generation (PTES, OWASP reporting standards).

Pentest skills operate strictly within written authorization and defined Rules of Engagement. They classify findings by OWASP category, CVSS v3.1, and MITRE ATT&CK, assess real-world exploitability, and produce sanitized remediation guidance — never weaponized exploit code. Active-testing skills set human_approval_required: true.

Subdomains covered by this domain:

  • Web application and API penetration testing (OWASP Top 10)
  • Penetration test report generation and CVSS-based risk rating

Skills Catalog

SkillSlugPrimary ToolMITRE Coverage
web-app-pentestpentest/web-app-pentestweb-app-pentest_tool.pyT1190, T1078, T1110, T1040, T1090
pentest-reportingpentest/pentest-reportingpentest-reporting_tool.py(Reporting / governance)

All skill paths are relative from the repository root as pentest/<slug>/.


Python Tools Reference

ToolPathPurposeKey Args
web-app-pentest_tool.pypentest/web-app-pentest/scripts/web-app-pentest_tool.pyEmits the skill status payload for orchestration--input, --output
pentest-reporting_tool.pypentest/pentest-reporting/scripts/pentest-reporting_tool.pyEmits the skill status payload for orchestration--input, --output

Authorization and Safety

  1. No active testing without confirmed written authorization (target, test window, Rules of Engagement).
  2. Out-of-scope systems are never tested, even when access is incidentally available.
  3. No weaponized exploit code in any output — proof-of-concept material is sanitized to reproduction steps.
  4. Active-testing findings set human_approval_required: true.
  5. Critical findings (RCE, auth bypass with data access) escalate immediately, ahead of the full report.

Domain Best Practices

  1. Validate scope and authorization as step zero. No reconnaissance, scanning, or exploitation guidance is produced until written authorization and the target list are confirmed. Missing authorization returns an advise output requesting the RoE.
  2. Rate every finding with CVSS v3.1 and a patch SLA. Severity drives the remediation deadline: Critical = 24h, High = 7d, Medium = 30d, Low = 90d.
  3. Deduplicate across sources before reporting. When findings arrive from multiple skills (web-app-pentest, network-pentest, red-team-operations, sast-dast-coordinator), merge duplicates and keep the highest severity.
  4. Translate technical severity into business impact for executives. A report read only by engineers has failed half its audience — pair each finding with the business consequence.
  5. Route findings onward. sast-dast-coordinator correlates code-level findings; compliance-mapping maps control gaps; findings-tracker tracks remediation to closure.

Related Domains

  • appsec-devsecops/ — sast-dast-coordinator correlates pentest findings with code-level findings
  • red-team/ — red-team-operations and safe-exploitation feed pentest-reporting
  • risk-compliance/ — compliance-mapping consumes findings for control-gap analysis

Path Reference

pentest/<slug>/
  README.md          -- skill overview and quick commands
  SKILL.md           -- full skill specification and methodology
  scripts/           -- executable Python tools
  references/        -- supporting documentation
  expected_outputs/  -- representative tool outputs for validation
  assets/            -- supporting data files

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.