Skip to content
Get packs

Pentest Output Standards

Formats penetration-test output: Chinese analysis, vulnerability reports, negative results, and a change ledger with rollback.

Source Download skill
Repository
AIPentest/CyberStrikeAI
Stars
8,323
Licence
Apache-2.0
Files
1 file

What it does

This is a set of formatting and reporting rules for an AI agent doing authorized penetration testing. Once installed, the agent writes its analysis in Chinese while keeping commands, code and CVE IDs in English, and records each step's reasoning and results in a fixed structure. Verified findings are written up with severity, affected versions, exploit steps, evidence, impact and fixes, and failed tests are logged so they are not repeated. It also keeps a ledger of every write-type change made to a target, with rollback commands.

Great for

  • Security testers running AI-assisted penetration tests who want consistent, auditable reports
  • Teams using the CyberStrikeAI platform that need a standard output format for findings and negative results
  • Red teams that must track every change they make to a target so it can be reverted

Skip it if

  • Anyone testing systems without written authorization, since the skill is built for offensive security work and provides no authorization check of its own
  • Users looking for general writing or reporting help outside penetration testing, because the rules are specific to pentest workflows and reference tools such as record_vulnerability and upsert_project_fact that only exist in the host platform

Try asking

  • “Write up the verified SQL injection finding on the staging host with severity, reproduction steps and fix advice”
  • “Summarise the current test state and next steps from the project facts so far”
  • “Log the webshell and config backup I created during the test and generate a rollback script”
Step-by-step guide: how to use Pentest Output StandardsInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    pentest-output-standards.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip pentest-output-standards.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip pentest-output-standards.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Pair it with the platform's fact and vulnerability tools, since the skill assumes they exist and does not work standalone
  • Review the change ledger and generated rollback script before running them, because the skill itself does not perform cleanup
  • Only use it on systems you are explicitly authorised to test

What's inside

  • SKILL.md

Redistributed unmodified from AIPentest/CyberStrikeAI under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.