agentleFS
zyphr-labszyphr-labs/bounty-harness/.claude/skills/impact-verifier/SKILL.md

impact-verifier

Skill1 starsChanged 3 months agoMITView on GitHub

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Verify: Run the specific verify workflow for the impact class.

    line 16
  2. Distinctive:

    All standard evidence artifacts present.

    line 45

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill 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
SKILL.md

Impact Verifier

Overview

The final gate before reporting. Converts candidate findings into confirmed bounty-grade reports only when impact is proven. Enforces evidence completeness, false-positive checks, impact classification, and report readiness scoring.

Quick Reference

  • Skill: impact-verifier
  • Version: 1.1.0
  • Bounded Context: ImpactContext
  • Required tools: python3, jq, curl
  • Risk tier: passive (validates, does not test)

Workflow Selection

  • Collect: collect-candidates from all skill output directories.
  • Classify: classify-impact to determine impact class.
  • Verify: Run the specific verify workflow for the impact class.
  • Gate: false-positive-gate and report-readiness before reporting.
  • Score: cvss4-score for CVSS 4.0 severity, then impact-narrative for the impact write-up.

Available Workflows

WorkflowPurpose
collect-candidatesAggregate candidate findings from all skill outputs.
classify-impactDetermine impact class for a candidate finding.
verify-data-exposureProve actual data exposure from unauthorized access.
verify-privilege-escalationProve role/permission boundary was broken.
verify-account-takeoverProve full account takeover path.
verify-tenant-breakProve cross-tenant data access.
false-positive-gateRun automated false-positive checks.
report-readinessScore finding readiness for submission.
cvss4-scoreCalculate CVSS 4.0 score, macrovector, and severity from a vector.
impact-narrativeGenerate impact narrative with attack scenario and PII detection.

Workflow Selection Notes

  • cvss4-score needs CVSS4_VECTOR in context (full vector string starting with CVSS:4.0/).
  • impact-narrative needs FINDING_DIR (evidence directory with response.txt) and IMPACT_CLASS (e.g. account_takeover, data_exposure, tenant_break, rce).
  • The narrative auto-detects sensitive fields (email, token, card_number, ...) in the captured response and strengthens the evidence section accordingly.

Evidence Required

  • Impact class confirmation with proof artifacts.
  • False-positive checklist completed.
  • Report readiness score >= 80.
  • All standard evidence artifacts present.
  • CVSS 4.0 vector string with metric breakdown (cvss4.json).
  • Impact narrative with attack scenario and observed data classes (impact.md).
  • Redact credentials, session cookies, bearer tokens, and auth headers from all collected evidence before reporting. Evidence stays local-only under $OUTDIR/impact-verifier and is never committed; sanitize requests/responses (gitignore evidence directories) so no secrets leak into reports.

References

  • Bugcrowd VRT 1.18
  • FIRST CVSS v4.0 specification (macrovector bands)
  • OWASP ASVS V4 (Access Control)
  • OWASP Reporting guidelines

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.