report-agent

Converts triage-approved security findings into HackerOne-format reports with CVSS scoring.

2|1|Updated Aug 28, 2026
One-click install
npx skills add https://github.com/lycheer1126/xs-bigdan --skill report-agent-lycheer1126
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: report-agent
Source: https://github.com/lycheer1126/xs-bigdan/tree/main/knowledge/agents/report
Command: npx skills add https://github.com/lycheer1126/xs-bigdan --skill report-agent-lycheer1126

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning confirmed vulnerability findings into professional, submission-ready reports is slow and inconsistent. This Skill standardizes the final phase of a security testing pipeline by converting triage-approved findings into structured HackerOne-format reports with consistent severity scoring and evidence chains. ## Core Features & Use Cases - HackerOne-Standard Reports: Generates reports with Summary, Steps to Reproduce, PoC, Impact, CVSS 3.1, and Mitigation sections following the platform's expected structure. - CVSS 3.1 Scoring: Produces vector strings and numeric scores with a severity reference table covering Critical through Low findings. - Evidence Chains: Outputs minimal PoC curl commands, annotated evidence references, and remediation guidance with CWE references. - Use Case: After a triage stage confirms an IDOR on /api/users/{id}, use this Skill to produce a complete report titled "IDOR on /api/users/{id} allows unauthorized access to user PII" with reproduction steps and a CVSS score ready for submission. ## Quick Start Generate a HackerOne-format report with CVSS scoring and PoC evidence for the triage-approved finding on the target endpoint.

Frequently Asked Questions about report-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a HackerOne vulnerability report?

Structure the report with a Summary, numbered Steps to Reproduce (3-15 steps), a Proof of Concept with full HTTP request and response, an Impact section, a CVSS 3.1 score, and Mitigation guidance. Titles follow the format "[Vuln Class] on [Endpoint] allows [Attacker] to [Impact]".

How do I calculate a CVSS 3.1 score for a finding?

Assign a severity band first: Critical (9.0-10.0) for RCE or mass data breach, High (7.0-8.9) for SQLi with extraction, Medium (4.0-6.9) for XSS or limited IDOR, and Low (0.1-3.9) for info disclosure. Then produce the full vector string alongside the numeric score.

What should a proof of concept include in a bug bounty report?

Include the complete HTTP request and response, an annotated screenshot, and a minimal curl command. Reproduction steps must be exact enough that anyone can verify the finding in under five minutes.

When does this report generation step run in a testing pipeline?

It runs as the final phase, only after findings pass triage approval. Unconfirmed or raw scanner output should not reach this stage, since the reports assume verified impact and reproducible evidence.

What are the limitations of automated vulnerability report generation?

Report quality depends entirely on the accuracy of the triage-approved input findings and their evidence. It does not verify exploitability itself, so weak or incomplete PoC evidence upstream produces weak reports.