five vitals

Diagnose any system against 5 universal structural signals. Get a scorecard (0-10), identify gaps, and receive targeted prescriptions. Works on software, AI agents, organizations, knowledge bases, and workflows.

aegis

Audit, fix, and harden any codebase for security vulnerabilities. STRIDE threat modeling, 300+ vulnerability patterns, 8 compliance frameworks, 5-dimension risk scoring, secure code transforms, and hardening configurations.

agent orchestration skill

Prevents context window explosion in multi-agent squads. Generates compact handoff artifacts (~379 tokens), enforces token budgets, and provides compaction-vs-handoff decision trees.

strata

Universal methodology for transforming raw ideas into state-of-the-art artifacts through 6 layers of independent adversarial compression. Adaptive depth calibrates intensity to complexity.

bvs lite

Free edition of Build-Verify-Seal. 6 mandatory architecture questions + basic XP Scorecard. Prevents coding without clarity. Upgrade to full BVS for module contracts, build order, seal criteria, and drift prevention.

visual-mind

Intelligent infographic generation skill. Classifies information topology, profiles audience trust, validates cognitive load, then generates SVG and tested image prompts. 20 layouts, 17 styles, CHI 2025 research.

build-verify-seal

Anti-hallucination engineering framework for Claude Code. 3-phase pipeline: Architecture (6 Questions) > Engineering (Build-Verify-Seal) > Verification (XP Scorecard). Enforces contracts, bottom-up build order, seal criteria, and drift prevention.

dendrite

Biomimetic debug methodology — 7-phase system for finding root causes through Bayesian elimination, information-theoretic test selection, and cross-module immunity.

mãe intuitiva crm

CRM inteligente para gestão de leads e agentes IA — Mãe Intuitiva

hr recruit ai

Smart recruitment toolkit — screens resumes, generates job descriptions, builds interview question banks, and drafts candidate communications. Bias-aware and compliance-conscious.

launch copy

Generates all launch copy assets from a brief — emails, landing page, ads, and social posts — using multi-master methodology.

radar lite — ai-ready website audit

RADAR Lite — 33-check website audit for AI visibility. Scores Foundation, Discoverability, Comprehension, Citability.

wiring doctor

Diagnoses and prescribes wiring in Claude Code systems. Reads the real wiring state, maps it against 6 invariants (structural, economic, reachability, propagation, boundary, semantic), scores fragility of each wire, and prescribes surgical corrections with explicit decision-tree reasoning and forward simulation.

algerian-wordpress-ecommerce-builder-pro

AI-powered WordPress & WooCommerce Builder optimized for Algeria, with multilingual support, COD setup, SEO, security, and performance optimization.

docforge

Master orchestrator for the docforge documentation lifecycle pack. Bootstrap, audit (DDS-1 drift score), generate (15 artifact types: README, ADR, runbook, postmortem, migration-guide, security-advisory…), release-prep, archive, refuse-with-reason. Enforces 5 anti-patterns (hedge soup, wall-of-text, doc theater, zombie doc, anonymous best-practice) and 5 editorial principles (audience-first, refuse-before-ship). Compliance reports for SOC2/GDPR/ISO27001.

hr performance

Assists with performance reviews, feedback writing, 360 synthesis, goal-setting (OKR/SMART), and individual development plans. Bias-aware with SBI feedback model.

heavy think

Execute heavy parallel reasoning followed by criterion-driven synthesis when single-pass thinking is insufficient. K independent trajectories + dual-mode deliberation (verification/synthesis/hybrid) + cascade detection. Empirically grounded in Wang et al. (2026).

docforge llms

Project canonical Markdown docs into AI-native projections — llms.txt (Howard 2024 spec), llms-full.txt (segmented), AGENTS.md (AAIF/Linux Foundation Dec 2025), CLAUDE.md (Anthropic verbatim, ≤200 lines). Includes MCP server scaffold (docs as typed resources), cache-aware payload ordering for 90% prompt-cache savings, RAG-friendly chunkability lint, audience-graded projections.

oficina atende

Generates complete WhatsApp customer service scripts for Brazilian auto repair shops and dealerships — from first message to confirmed work order. Includes lead qualification, scheduling, confirmation and no-show prevention.

oficina fideliza

Generates complete automated post-service sequences for Brazilian auto repair shops and dealerships — NPS, future service reminders, no-show recovery, and seasonal alerts. Turns one-time customers into recurring clients.