Multi-agent security vulnerability analysis with adversarial verification and ICD 203 analytic standards. Orchestrates 5 parallel finder agents, cross-model adversarial verification (Claude + Codex), and deterministic validation to analyze vulnerability reports with CWE-specific procedures, confirmation bias mitigation, and structured evidence quality assessment. Includes a `--develop-fix` mode (Rust-first v1) that, after confirming a concrete target, authors validated regression tests + a minimum fix on a branch and emits a human-gated **candidate patch** — it never confirms its own fix; the mandatory `--verify-fix` gate and a human merge do. Use when receiving vulnerability reports, security disclosures, or bug bounty submissions, when assessing security issues, or when authoring a candidate fix for a confirmed finding.
Preservation-first forensic evidence collection and analysis for Vercel security incidents (OAuth supply-chain compromise, env-var exposure, audit-log triage). Produces a frozen evidence directory, an 8-section findings report, and a rotation-worklist CSV for downstream rotation tooling. Strictly read-only. Never rotates, revokes, deletes, or redeploys. The banned-ops list is absolute. Rotation is handed off to subinium/vercel-incident-toolkit Flow C or codyhxyz/metapod-harden `/rotate-vercel-env <KEY>` using this skill's CSV as input. Use this skill when: (1) Vercel publishes a security incident and you need to assess exposure across your team / account / linked GitHub org. (2) A customer suspects env-var exposure and wants an evidence bundle plus prioritized rotation worklist by lunch Monday. (3) Audit-log triage across Vercel activity + GitHub audit log is needed without contaminating the very log under investigation. (4) Forensic handoff to a rotation tool is required — this skill produces
Vanta compliance platform operations — posture analysis, audit readiness, vulnerability management, personnel compliance, and flexible reporting. Complements the official vanta-mcp-plugin with analysis workflows, direct API access for write operations, and reporting capabilities not available in the Vanta UI. 50% compliance analysis/reporting, 30% API operations, 20% workflow orchestration. Use this skill when users need to: (1) Assess compliance posture across frameworks (gap analysis, control coverage, cross-framework overlap) (2) Prepare for audits (readiness checklist, missing evidence, policy expiry, personnel compliance) (3) Track and triage vulnerabilities with SLA awareness (approaching deadlines, missed SLAs, severity breakdown) (4) Monitor personnel compliance (overdue training, policy acceptance, deactivated personnel in scope) (5) Generate compliance reports and executive summaries (posture, readiness, vulnerability SLA, personnel, custom) (6) Perform bulk low-risk operations
Three-layer Stop hook that detects and blocks work-skipping rationalizations by Claude Code agents. Prevents agents from unilaterally deciding to skip assigned work, bypass mandatory processes, or cite unverified context constraints as justification for shortcuts. Layer 1: Deterministic regex detection of known laziness phrases (Tiers 1-2) Layer 2: Haiku-evaluated internal contradiction detection (Tier 4) Layer 3: Context-aware agent verification — auto-activates when plans/tasks exist (Tier 5) This is a passive hook — it activates automatically when the plugin is enabled. No slash command needed. The hook fires on every Stop event and only blocks when laziness patterns are detected.
Docker Sandboxes (sbx CLI) — run AI coding agents in isolated microVM environments with credential proxying, network policy enforcement, and custom templates. Covers Claude Code, Codex, Copilot, and Gemini agents. Compatible with Rancher Desktop (Docker Desktop not required). Use this skill when users need to: (1) Install, configure, or authenticate with the sbx CLI (2) Create, run, stop, or remove sandboxed agent sessions (3) Configure credentials, secrets, or API keys for sandboxed agents (4) Understand the security model (microVM, proxy, network policies) (5) Build custom sandbox templates or customize environments (6) Troubleshoot sandbox issues (clock drift, port forwarding, connectivity) (7) Use branch mode, multi-workspace, or reconnection workflows (8) Set up specific agents (Claude Code, Codex, Copilot, Gemini) in sandboxes (9) Use 1Password CLI (op) for zero-disk-footprint secret injection
Codebase flow analysis for dependency visualization, impact assessment, and health scoring. Use PROACTIVELY when users need: (1) Dependency/import analysis ("what imports this?", "dependency graph", "module relationships") (2) Blast radius/impact analysis ("if I change X", "what's affected", "impact of modifying") (3) Code health scoring ("codebase health", "health grade", "technical debt score") (4) Execution flow tracing ("trace through", "call path", "how does data flow") (5) Dead code detection ("unused exports", "safe to delete", "orphan code") (6) Comprehensive analysis ("full analysis", "analyze this codebase") NOTE: For security scanning, defer to security-sentinel. For design patterns, defer to pattern-recognition-specialist. For architecture compliance, defer to architecture-strategist.
World-expert deep technical research agent for AI-enabled software development. Use PROACTIVELY when users need: (1) Deep research on AI/ML development topics (RAG, agents, LLMs, embeddings, vector DBs, prompt engineering, fine-tuning) (2) Technical consultation on AI architectures, tool selection, or implementation approaches (3) Implementation guidance with production-ready patterns and best practices (4) Comparative analysis of AI frameworks, models, or services (5) Current state-of-the-art analysis with authoritative citations
Discover trending AI tools, news, and insights from influential developers and AI advocates on Twitter/X using Bird CLI. Use PROACTIVELY when users need: (1) AI-related tweets, tools, or discussions from Twitter (2) What influential AI developers are currently talking about (3) AI news and trends from social media (4) Developer sentiment on AI tools and frameworks (5) New AI projects being shared on Twitter This is a READ-ONLY skill - no posting, liking, or account modifications.