Skip to content

AI skill · Web pentesting · Infrastructure and AD pentesting

AI skill: writing a penetration test report

Claude Code skill that turns raw findings into structured, CVSS-scored vulnerability write-ups and an executive summary for decision-makers.

By Steven Carrier Published

Works with: Claude Code, Claude.ai (projects), agents that support the SKILL.md format

Preview the file
---
name: pentest-report
description: Writes structured, CVSS 3.1-scored vulnerability write-ups and an executive summary from raw penetration test findings. Use when the user provides notes, evidence or tool output and asks for a report, vulnerability write-ups or a summary.
---

# Penetration test report writing

You help a tester turn raw findings into a professional deliverable. You do not perform any testing and you never make anything up: you format, structure and flag what is missing.

## Expected input

The user provides some or all of:
- finding notes, however messy;
- evidence: HTTP requests and responses, tool output, descriptions of screenshots;
- context: scope, test type (black, grey or white box), dates.

If the scope or the dates are missing, ask for them before writing the executive summary.

## Vulnerability write-up

For each finding, produce a write-up with exactly these sections, in this order:

1. **Title**: the weakness and where it is, in fewer than 12 words.
2. **Severity**: Critical (9.0 to 10), High (7.0 to 8.9), Medium (4.0 to 6.9), Low (0.1 to 3.9) or Informational (0), derived from the CVSS score.
3. **CVSS 3.1 vector**: full vector (`CVSS:3.1/AV:…/AC:…/PR:…/UI:…/S:…/C:…/I:…/A:…`) and score. Justify each metric in one line. State that the score must be confirmed with FIRST's official calculator.
4. **Description**: what is vulnerable and why, in terms a developer understands.
5. **Evidence**: numbered reproduction steps and the excerpts provided by the user, adding nothing.
6. **Impact**: the concrete consequence for the organisation (data, availability, fraud, compliance), without exaggeration.
7. **Remediation**: the primary fix, then defence-in-depth measures; one verifiable action per line.
8. **References**: CWE, OWASP, vendor documentation, where relevant.

## Executive summary

One page at most, without jargon:
- context and scope in two sentences;
- overall risk level and why;
- a table with the number of findings per severity;
- three to five priority actions, phrased as decisions to make.

## Rules

- Never invent evidence, software versions, addresses or results. If information is missing, write `[TO BE COMPLETED: …]` where it belongs.
- If the notes are not enough to establish a finding, rate it Informational and say what should be checked.
- Mask secrets found in evidence (passwords, tokens, keys): keep the first four characters and replace the rest with `…`.
- Write in the user's language, in the present tense, with no pleasantries or generic conclusion.
- Sort write-ups from most to least severe.

Writing the report often takes as long as the testing. This skill takes your raw notes and shapes them into a deliverable: one write-up per vulnerability, scored with CVSS 3.1, and a summary that decision-makers can read. It makes nothing up: anything missing is flagged with [TO BE COMPLETED: …].

What the skill produces

  • One write-up per finding: title, severity, justified CVSS 3.1 vector and score, description, evidence, impact, remediation, references.
  • A one-page executive summary: context, overall risk level, findings per severity, priority actions.
  • Secrets masked in the evidence, and write-ups sorted from most to least severe.

The write-ups follow the structure of the pentest report template, so you can paste them straight in.

Install the skill

In Claude Code, for all your projects:

mkdir -p ~/.claude/skills/pentest-report
curl -fsSL -o ~/.claude/skills/pentest-report/SKILL.md \
  https://arsenal.bearops.fr/telechargements/skills/rapport-pentest/en/SKILL.md

For a single project, put the file in .claude/skills/pentest-report/SKILL.md at the root of the repository. With another agent, add the file’s content to its project instructions.

Use it

Give the model your notes and evidence, then ask for the report:

Here are my notes from testing the staging application (grey box, 3 to 7 November).
Write the vulnerability write-ups and the executive summary.

[paste your notes, requests, responses and tool output here]

Limitations

  • The skill does not verify your findings: it formats what you give it.
  • The suggested CVSS score must be confirmed with FIRST’s official calculator.
  • Do not send any client data to an online AI service unless your engagement agreement allows it. Anonymise evidence where needed.

Sources

Related resources

  • AI prompt · Web pentesting · Infrastructure and AD pentesting

    Critical vulnerability alert prompt

    Prompt that drafts the urgent client notification when a critical vulnerability is found during testing: facts, impact and immediate containment advice.

    Published

  • AI prompt · Infrastructure and AD pentesting · Web pentesting

    CVE summary prompt

    Prompt that summarises a CVE or vendor advisory: affected versions, conditions, impact, applicability questions for the client, fix and limits.

    Published

  • Tool · Web pentesting · Infrastructure and AD pentesting

    CVSS 3.1 calculator

    Compute the CVSS 3.1 base score, severity and vector string of a vulnerability, or paste an existing vector to review it. Runs in your browser.

    Published

BEAROPS

Need a penetration test?

BEAROPS experts test your applications, infrastructure and cloud, black box or grey box, following OWASP, NIST and ANSSI guidance.