agentleFS
fatihkanfatihkan/badi/.claude/skills-vault/security-check/SKILL.md

security-check

Skill8 starsChanged 3 months agoMITView on GitHub

Comprehensive AI-powered security scanning suite with 48 skills covering OWASP Top 10, 7 language-specific deep scanners (Go, TypeScript, Python, PHP, Rust, Java, C#), supply chain analysis, infrastructure-as-code scanning, and 3000+ checklist items. Use when you need to run a security audit, find vulnerabilities, scan a PR for security issues, or perform a penetration test on a codebase.

33 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use learn and code-review.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • genomewalkerSkill

    learn

    Learn from web sources, generate SKILL.md, and store key patterns in soul memory. Trigger keywords: learn, skill-gen, web-learn, documentation, tutorial.

    Also in 3 projects that use this

  • GoogleSkill

    code-review

    Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.

    Also in 5 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 25 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No binaries, no dependencies, no CI pipeline changes.

    line 31
  2. Distinctive:

    After installation, open your AI assistant and say:

    line 35
  3. Very specific:

    After scanning, a security-report/ directory is created containing:

    line 74

“Common” is measured: kept by at least 6 in 10 of the 25 edited versions of this file we could read. “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

security-check

Your AI Becomes a Security Team. Every Language. Every Layer. Zero Tools.

Entry points (v1.31.0+):

  • /security-review — Anthropic native command (Claude Code 2.1.140+, AI semantic)
  • badi security baseline — deterministic baseline (secret-scan + audit)
  • badi security triage — /security-review report severity filtering
  • This skill: sc-orchestrator 4-phase pipeline (Recon → Hunt → Verify → Report)

What This Skill Does

security-check transforms your AI coding assistant into a comprehensive security scanning team. It runs a 4-phase pipeline — Recon → Hunt → Verify → Report — entirely through natural language. No binaries, no dependencies, no CI pipeline changes.

Quick Start

After installation, open your AI assistant and say:

  • "run security check" — Full security audit
  • "scan diff" — PR/diff-level incremental scan
  • "scan for vulnerabilities" — Same as full scan

What's Included

48 Security Skills

CategoryCountSkills
Core Pipeline6Orchestrator, Recon, Dependency Audit, Verifier, Report, Diff Report
Injection9SQLi, NoSQLi, GraphQL, XSS, SSTI, XXE, LDAP, CMDi, Header Injection
Code Execution2RCE, Deserialization
Access Control4Auth, AuthZ, Privilege Escalation, Session
Data Exposure3Secrets, Data Exposure, Crypto
Server-Side4SSRF, Path Traversal, File Upload, Open Redirect
Client-Side4CSRF, CORS, Clickjacking, WebSocket
Logic & Design3Business Logic, Race Conditions, Mass Assignment
API Security3API Security, Rate Limiting, JWT
Infrastructure3IaC, Docker, CI/CD
Language Scanners7Go, TypeScript, Python, PHP, Rust, Java, C#

10 Security Checklists (3000+ items)

Each language scanner includes a 400+ item checklist with specific CWE references.

4-Phase Pipeline

Phase 1: RECON        → Architecture mapping, tech stack detection
Phase 2: HUNT         → 40+ vulnerability skills run in parallel
Phase 3: VERIFY       → False positive elimination, confidence scoring
Phase 4: REPORT       → CVSS severity, remediation roadmap

Output

After scanning, a security-report/ directory is created containing:

  • SECURITY-REPORT.md — Final consolidated report
  • architecture.md — Codebase architecture map
  • dependency-audit.md — Supply chain analysis
  • verified-findings.md — Findings after false positive elimination

Harness-Compatible Artifact Chain (v1.34+)

In addition to security-report/, the pipeline emits a file-contract chain at the project root using the artifact names of Anthropic's defending-code-reference-harness:

THREAT_MODEL.md  →  VULN-FINDINGS.json/.md  →  TRIAGE.json/.md
(pentest-threat-model)   (Phase 2: hunt)         (Phase 3: verify)
  • VULN-FINDINGS.json — raw findings after Phase 2 (schema in sc-orchestrator); confidence is null by design — the producer never judges confidence
  • TRIAGE.json — verdicts after Phase 3 (TRUE_POSITIVE / FALSE_POSITIVE / CANNOT_VERIFY, dedupe links, 0-10 one-decimal confidence; mapping in sc-verifier)
  • THREAT_MODEL.md — optional chain head from the pentest-threat-model skill; when present, Phase 1 recon uses its entry points and trust boundaries for scoping

Interop is outbound and name-level: badi emits the harness's artifact names and verdict vocabulary so output is structurally familiar to harness-side tooling. It does not guarantee field-value ingestion — badi diverges by design (confidence: null at the producer; duplicate_of side field instead of a duplicate verdict). badi does not require the harness to run. This is a simplified adaptation — the upstream autonomous pipeline's multi-vote verification and sandboxed (gVisor + ASAN) execution stages are not reproduced here.

Recommend gitignoring the generated VULN-FINDINGS.* and TRIAGE.* files; THREAT_MODEL.md is a durable design document and is usually committed.

Provenance

The artifact-chain contract (THREAT_MODEL.md → VULN-FINDINGS → TRIAGE filenames and field conventions) is adapted from anthropics/defending-code-reference-harness (Apache-2.0, Copyright Anthropic PBC). Only the artifact names and field conventions are adapted; no upstream source code is copied. The skill text, the sc-* pipeline, and its verification methodology are independently authored for badi (MIT). This Provenance notice and the upstream link satisfy the Apache-2.0 attribution requirement.

More Information

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.