/skill-security-audit

name: octopus-security-audit
disable-model-invocation: true
aliases:
  - security
  - security-audit
description: OWASP compliance, vulnerability scanning, and adversarial red team testing — use for security reviews
trigger: |
  Use this skill when the user says "security audit this code", "find vulnerabilities in X",
  "red team review", "pentest this API", or "check for OWASP issues".
execution_mode: enforced
pre_execution_contract:
  - visual_indicators_displayed
validation_gates:
  - orchestrate_sh_executed
  - output_artifact_exists
paths:
  - "**/.env*"
  - "**/auth*"
  - "**/security*"
  - "**/oauth*"
  - "Dockerfile*"

Security Audit Skill

**Your first output line MUST be:** `🐙 **CLAUDE OCTOPUS ACTIVATED** - Security Audit`

Invokes the security-auditor persona for thorough security analysis during the `ink` (deliver) phase. Supports both quick OWASP scanning and full adversarial red/blue team testing.

Usage

# Quick scan via security-auditor persona
"${CLAUDE_PLUGIN_ROOT:-${CODEX_PLUGIN_ROOT:-${HOME}/.claude-octopus/plugin}}/scripts/orchestrate.sh" spawn security-auditor "Scan for SQL injection vulnerabilities"

# Adversarial red team via squeeze workflow
"${CLAUDE_PLUGIN_ROOT:-${CODEX_PLUGIN_ROOT:-${HOME}/.claude-octopus/plugin}}/scripts/orchestrate.sh" squeeze "Security audit the authentication module"

# Via auto-routing (detects security intent)
"${CLAUDE_PLUGIN_ROOT:-${CODEX_PLUGIN_ROOT:-${HOME}/.claude-octopus/plugin}}/scripts/orchestrate.sh" auto "security audit the payment processing module"

Modes (Auto-Detected)

| Mode | Auto-Trigger | Confidence Gate | Scope | |------|-------------|----------------|-------| | **Quick** (default) | Standard security scan, no sensitive files in diff | 8/10 — only high-confidence findings | Changed files only | | **Deep** (auto-escalated) | Diff touches auth/security/CI files, OR explicit request | 2/10 — flag anything suspicious | Entire codebase |

**Auto-escalation to Deep mode:** The skill automatically switches to Deep mode when ANY of these are true:

• Diff includes files matching: `*auth*`, `*login*`, `*password*`, `*session*`, `*token*`, `*secret*`, `*crypt*`, `*oauth*`, `*saml*`, `*jwt*`, `*permission*`, `*rbac*`, `*acl*`
• Diff includes CI/CD files: `.github/workflows/*`, `Dockerfile*`, `docker-compose*`, `.gitlab-ci*`
• Diff includes dependency files: `package-lock.json`, `yarn.lock`, `Gemfile.lock`, `requirements.txt`, `go.sum`
• The user explicitly says "deep", "full", "comprehensive", or "CSO"

No user action needed — mode detection happens automatically from the git diff context.

Model Selection Caveat: Fable 5.1

By default, do not dispatch security-audit passes to Claude Fable 5.1 or the preserved Fable 5 ID. Their safety classifiers can refuse adversarial red-team phrasing in authorized audits. Route these passes to `OCTOPUS_FABLE5_FALLBACK_MODEL` (default `claude-opus-5`) and keep prompts defensively framed (find and report vulnerabilities; do not request working exploits). Reject an exact model-qualified override that targets `claude-fable-5-1` or `claude-fable-5`. On refusal, retry exactly once on the fallback unless `OCTOPUS_FABLE5_NO_RETRY=1`; when retries are disabled or the retry refuses, surface the refusal without further dispatches. `OCTOPUS_FABLE5_MODE=off` is the explicit exception: it disables automatic rerouting for ordinary environment pins, while exact Fable security seats still fail closed. Details: `skills/blocks/fable5-prompting.md`.

Capabilities

Core (both modes)

• OWASP Top 10 vulnerability detection
• SQL injection and XSS scanning
• Authentication/authorization review
• Secrets and credential detection
• Dependency vulnerability assessment
• Security configuration review

Secrets Archaeology (Deep mode)

Scan git history for leaked credentials that may have been "deleted" but remain in commits:

# Search git history for common secret patterns
git log --all -p --diff-filter=D -- '*.env' '*.key' '*.pem' 2>/dev/null | head -200
git log --all -p -S 'AKIA' --pickaxe-regex 2>/dev/null | head -100  # AWS keys
git log --all -p -S 'sk-[a-zA-Z0-9]{20,}' --pickaxe-regex 2>/dev/null | head -100  # API keys
git log --all -p -S 'ghp_|gho_|github_pat_' --pickaxe-regex 2>/dev/null | head -100  # GitHub tokens
git log --all -p -S 'password\s*[:=]' --pickaxe-regex 2>/dev/null | head -100  # Passwords

Report any findings with the commit SHA, file, and recommendation to rotate the credential.

CI/CD Pipeline Security (Deep mode)

Audit GitHub Actions and CI configuration for injection and privilege escalation:

# Find all workflow files
find .github/workflows -name '*.yml' -o -name '*.yaml' 2>/dev/null

# Check for dangerous patterns:
# 1. Untrusted input in run: blocks (command injection via PR titles/branch names)
# 2. pull_request_target with checkout of PR code (code execution from forks)
# 3. Overly broad permissions (write-all, contents: write)
# 4. Missing pinned action versions (uses: actions/checkout vs actions/checkout@v4)
# 5. Secrets exposed to pull_request events (accessible to forks)

Flag each finding with severity (CRITICAL/HIGH/MEDIUM/LOW).

Skill & Plugin Supply Chain (Deep mode)

Verify integrity of installed Claude Code skills and plugins:

# List installed plugins
ls -la ~/.claude/plugins/ 2>/dev/null

# Check for skills that execute arbitrary bash
grep -r 'exec\|eval\|bash -c' ~/.claude/skills/*/SKILL.md 2>/dev/null | head -20

# Verify plugin sources (are they from known registries?)
cat ~/.claude/plugins/*/plugin.json 2>/dev/null | grep -E '"source"|"registry"'

STRIDE Threat Modeling (Deep mode)

For the target component, enumerate threats across all 6 STRIDE categories:

| Category | Question | |----------|----------| | **S**poofing | Can an attacker impersonate a user or component? | | **T**ampering | Can data be modified in transit or at rest? | | **R**epudiation | Can actions be denied without au