Mcp Builder
Builds MCP (Model Context Protocol) servers for API integration. Guides the full development lifecycle from research to evaluation.
Counterfactual
Generate structured alternate timelines for a past decision log entry and write them as a `reflection` entry under your decision wiki's counterfactuals/ directory. V1 ships Shell 2x2 fork enumeration (default) + Wachter near/medium/far fallback, single-pass simulation, one adversarial grill pass, mermaid quadrantChart visualization. Counterfactuals are PLACES with TRACK RECORDS — not thoughts. Wr…
Pdf Generation
Generates PDFs using Typst templates, handles print layouts, and processes images with Sharp.
Prompt Grounding
Audits a prompt against the actual codebase, docs, and strategy files — separates fact from invention, finds what's already built, identifies the real gap, and rewrites the prompt grounded in documented reality. Use when a prompt makes claims about your systems, products, or strategy that need verification before sending to a model.
Sprint Architect
Converts strategy documents, issue lists, or audits into dependency-aware agent-team execution boards. Produces wave-parallelized sprint boards with decision gates, capacity models, quality gates, and dispatcher protocols.
Data Analyst Agent
Build AI agents that analyze datasets, create visualizations, and extract insights through autonomous code execution.
Database Migrations
Modifies database schemas safely with migration strategies, rollback procedures, and zero-downtime deployments using Drizzle ORM.
Breakdown
Produce a multi-altitude explainer of a body of work — what it is, why it matters, where it sits, and how it was learned. Trigger when the user says "break it down", "explain what we did/built", "what was this", "give me the story", "zoom out", "explain at 100k/50k/30k/10k ft", "one-sentence version", "package this up", "what's the shape of this", or asks for ASCII C4 diagrams of work just comple…
Adversarial Research
Deep research where agents adversarially fact-check each other's findings instead of one agent confidently citing sources it never verified. Scrapes claims from X/web/GitHub, then spins up independent verifier agents that try to REFUTE each claim, labeling them survives / cut / unconfirmed with source links. Use for "fact-check these claims", "what holds up vs hype", "peer-review this research",…
Scratchpad
Compaction-resilient session notepad. Save working context that survives long conversations. Three tiers — pinned (always loaded), working (session), manual (permanent).
Package Research
Researches packages, dependencies, and APIs before integration. Leverages package-search MCP to ground responses in real source code.
Deep Recon
Multi-agent research swarm that finds existing solutions before you write a single line of code. Prevents reinventing the wheel.
Model Migration
Generate a personalized "how to use this new model" report by mining ALL your local Claude Code sessions, then comparing your actual usage patterns against the new model's documented behavior. Use when a new Claude/Codex/Gemini model ships (e.g. "how should I use Opus 5", "model migration", "what changes with the new model", "tailor my workflow to <model>"). Produces an evidence-backed report gro…
Fullstack Agent
Enables AI agents to create complete web applications with file system operations, code execution, template sandboxes, and live preview through sandbox URL hosting.
Ship
Fix-until-clean loop then commit and push. Use when ready to ship code — runs tests, lint, typecheck, fixes any issues, commits, and pushes.
D3js Visualization
Creates data visualizations with D3.js including custom charts, network diagrams, geographic maps, and publication-quality interactive graphics.
Naming The Machine
Scan skills, tools, workflows, and infrastructure to identify and name emergent expressions of value — products, methodologies, brands, or IP assets hidden in existing systems. Triggers on "what have I built," "find products in my systems," "audit for marketable expressions," or "name everything.
Executing Plans
Implements written plans through batched execution with review checkpoints. Stops on blockers and asks for clarification rather than guessing.
Frontend Design
Builds web components, pages, applications, landing pages, and styled web UIs. Creates distinctive, production-grade frontend interfaces that avoid generic AI aesthetics.
Webapp Testing
Provides Playwright patterns and best practices for web application testing. Covers DOM inspection, screenshot capture, form interaction, and API mocking. Works with the Playwright plugin for browser automation.
Requesting Code Review
Structures code review requests with context, commit hashes, and requirements. Facilitates early and frequent review cycles.
Design Taste
Senior UI/UX Engineer. Architect digital interfaces overriding default LLM biases. Enforces metric-based rules, strict component architecture, CSS hardware acceleration, and balanced design engineering.
Compound Engineering
Connected analyze-plan-execute-learn-repeat cycle with persistent memory. The human-in-the-loop version of Ralph for improving existing projects.
Test And Fix
Run tests and auto-fix failures. Standalone test runner that diagnoses and repairs failing tests without manual intervention.
Prompt Engineer
Turn a rough natural-language ask into a high-quality, Anthropic-spec prompt. Use when the user wants to write, engineer, improve, tighten, or harden a prompt — "write me a prompt for X", "make this prompt better", "rewrite my prompt to spec", "engineer this prompt", "prompt for <task>", or /prompt. ALWAYS reads references/anthropic-prompt-engineering.md (the codified current Anthropic guidance)…
Web Asset Generator
Creates web assets including favicons, app icons, Open Graph images, social media cards, and other web graphics from source images or designs.
Kickoff
Full workflow orchestrator. Use when starting substantial work that benefits from research, planning, and structured execution. Chains recon → teardown → compound → execute → learn.
Grill
Adversarial code review — challenge your code harshly, find what will break in production. Inspired by Boris Cherny's setup.
Verification Before Completion
Mandates fresh verification evidence before any completion claims. Prevents assumptions by requiring proof that work is actually complete and correct.
E2b Sandbox
Secure cloud sandbox execution for AI coding agents using E2B. Isolated containers for safe untrusted code execution.
Critique
Single-pass 3-agent critique for ANY work product — plans, code, designs, architecture. The general-purpose refinement tool.
Context Budget
Audit token consumption across all loaded components — skills, agents, MCP tools, rules, hooks
Writing Plans
Creates detailed implementation plans for multi-step tasks with exact file paths, code examples, and test commands.
Slack Gif Creator
Creates animated GIFs optimized for Slack including emoji animations and message GIFs with proper sizing and frame rates.
C4 Architect
Generates and maintains C4 architecture diagrams (Context, Container, Component) as Mermaid in docs/architecture/. Triggers on /c4, or when significant structural changes are committed (new services, changed dependencies, new DB tables, new API routes). Auto-detects project from cwd.
Whatdidimiss
After a chunk of work is done, surfaces what the model noticed but was never asked to do — adjacent tasks, unraised risks, natural follow-ups, and contradictions with the original plan. Opt-in only; the model reviews the work it just completed and the context it gathered, then returns a tight ranked list, each item with a one-line "worth doing? (y/n)". Triggers on "/whatdidimiss", "what did I mis…
Canvas Design
Creates visual art in PDF and PNG formats covering design philosophies, aesthetic movements, and museum-quality graphics. Communicates through form, space, and color rather than text.
Techdebt
End-of-session tech debt cleanup. Scans for TODOs, dead code, unused imports, type issues, and quick wins before you log off.
Ecosystem Audit
Audit your entire .claude ecosystem (skills, agents, slash commands, hooks, rules, memory) for duplicates, overlapping capabilities, stale one-off demos, contradictory rules, and context bloat. Use ~monthly or when things feel cluttered — "audit my .claude", "find duplicate skills", "what's stale", "ecosystem audit", "clean up my agents", "find contradictory rules". Returns a low/medium/high prio…
Reverse Engineer To Learn
Studies how existing apps solve problems to inform new builds. Performs strategic teardowns for learning, scoping features, and finding packages. Reverse-engineers macOS app bundles, npm/pypi packages, GitHub repos, and web apps (via browser), then applies the strategic questions. Triggers — "teardown", "reverse engineer", "how does X work", "study this app/package".
Review Changes
Analyze uncommitted changes before committing. Quick pre-commit review of what you're about to ship.
Visual Verdict
Structured visual QA verdict for screenshot-to-reference comparisons. Returns JSON scores driving edit iterations. Use after any UI implementation to verify visual fidelity.
Threat Model
Repository-grounded threat modeling. Enumerates trust boundaries, assets, attacker capabilities, and abuse paths tied to actual code. Not generic OWASP checklists — evidence-backed claims with file/line references.
Api Integration
Integrates with external APIs including REST, GraphQL, and webhooks. Covers authentication patterns, error handling, rate limiting, and type-safe clients.
Coding Agents
Core patterns for building AI coding agents with tool calling, agent loops, and autonomous code execution — plus expert guidance on agent frameworks and production patterns. Use for AI agent development, agent architecture, framework selection, and production agent considerations.
Algorithmic Art
Creates generative and computational art using p5.js with seeded randomness, particle systems, and interactive exploration. Covers creative coding and procedural aesthetics.
Redesign
Upgrades existing websites and apps to premium quality. Audits current design, identifies generic AI patterns, and applies high-end design standards without breaking functionality. Works with any CSS framework or vanilla CSS.