Forces a design-first discussion before building features, components, or any new plan.

Analyzes a failed superpowers session and drafts a bug report for the maintainers.

Dispatches independent tasks to parallel agents, each working without shared state or sequential dependencies.

Lets the agent implement a written plan inline in the current session, without subagents.

Guides integrating finished, fully tested work by choosing merge, PR, or cleanup options.

Makes agents verify review feedback technically before implementing suggestions, not blindly agreeing.

Prompts a review of completed work against requirements before merging or opening a PR.

Executes independent plan tasks through subagents within the current session.

Enforces a root-cause investigation before any fix is proposed for bugs or test failures.

Enforces writing a failing test before implementation code for features and bugfixes.

Creates an isolated git worktree workspace for feature work or plan execution when needed.

Teaches the agent to find and invoke relevant skills before responding to any message.

Requires running verification commands and showing output before claiming work is complete or passing.

Turns a spec or multi-step requirements into a written implementation plan before coding starts.

Guides creating, editing, and testing skills before they are deployed to agents.

Routes you to the right mattpocock skill or flow for your current situation.

Pursue a long-running goal in one session by coordinating multiple subagents.

Hand the current conversation to a fresh background agent that continues the work immediately.

Reviews changes against repo coding standards and the originating spec, using parallel sub-agents.

Gives a shared deep-module vocabulary for designing interfaces, seams, and more testable code.

Runs a structured loop to diagnose hard bugs, throwing errors, and performance regressions.

Helps build and sharpen domain terminology, glossaries, and architecture decision records.

Block destructive git commands like push, reset --hard, and branch -D via Claude Code hooks.

Get relentlessly interviewed to expose gaps and sharpen a plan or design.

Stress-test a plan or design through relentless questioning while capturing ADRs and glossary terms.

Stress-test your plan, decision, or idea with relentless pointed questioning.

Compact the current conversation into a handoff document for another agent to resume.

Build a piece of work in code from a spec or ticket set.

Implement code from /to-spec and /to-tickets outputs.

Find codebase deepening opportunities, review them in an HTML report, then refine one.

Interview me about specs for workflows I want to build in this workspace.

Replace unsafe as type assertions in tests with Shoehorn for partial test data.

Write a pull request body that describes changes clearly for reviewers.

Build a throwaway prototype to test a state model or explore UI design.

Investigate questions in primary sources and save findings as a Markdown file.

Review a coding session to capture lessons and improve future work.

Generate lint-passing exercise directories with sections, problems, solutions, and explainers.

Configure a repo's issue tracker, triage labels, and domain docs for engineering skills.

Add Husky pre-commit hooks running Prettier, type checks, and tests on commit.

Configure dependency-cruiser so TypeScript packages expose only their entry-point modules.

Build features or fix bugs test-first using red-green-refactor cycles.

Teach you a new skill or concept directly inside your current workspace.

Turn an unanswerable decision into a questionnaire for someone else to complete.

Turn the current conversation into a published spec by synthesizing it, without interviewing.

Split a plan or conversation into tracer-bullet tickets with blocking edges, published to a tracker.

Route issues and external PRs through triage states and write agent-ready briefs.

Make the agent re-pitch a message that failed to land clearly.

Map large multi-session work as decision tickets and resolve them until the path is clear.

Generate an interactive bash wizard that guides humans through steps only they can perform.

Assemble raw material into a beat-by-beat narrative, grounding terms before relying on them.

Write effective skills, AGENTS.md, and CLAUDE.md files that agents follow reliably.

Mine raw fragments of ideas before any structure exists.

Turn raw material into a finished article, shaping it paragraph by paragraph.

Design, implement, and audit WCAG 2.2 AA accessibility, including ARIA and iOS/Android traits.

Audit a 12-layer LLM agent stack for regressions, memory pollution, tool misuse, and hidden loops.

Benchmark coding agents like Claude Code, Aider, and Codex head-to-head on pass rate, cost, and time.

Design agent action spaces, tool definitions, and observation formats to improve task completion rates.

Diagnose AI agent failures with structured self-debugging, containment, recovery, and introspection reports.

Add x402 payments to AI agents with per-task budgets, spend controls, and non-custodial wallets across chains.

Trim ECC to a repo's needed skills by sorting them into daily and library buckets.

Operate as an agentic engineer with eval-first execution, task decomposition, and cost-aware model routing.

Build a persistent multi-agent operating system in Claude Code with specialists, commands, memory, and scheduling.

AIエージェントが大量の実装を生成するチームの開発とレビューの運用を設計する。

DB不要のサンドボックスAPIテストで、AIが書いたコードのリグレッションを自動検出する。

AndroidとKotlin Multiplatformでモジュール構成、UseCase、Repositoryのクリーンアーキテクチャを設計する。

シグナルやフォーム、ルーティングを含むAngularコードを生成し、設計を助言する。

既存の統合パターンに沿って、新しいAPIコネクターやプロバイダーを追加する。

Design REST APIs with resource naming, status codes, pagination, versioning, and rate limiting.

REST APIの命名、ステータスコード、ページネーション、エラー応答を統一的に設計する。

コーディング中の設計判断を検出し、文脈や代替案を含むADRとして記録する。

提供された例やブランドの声に沿って、記事やガイドなどの長文を執筆する。

ジョブやフック、MCPサーバーの稼働状況と冗長性を証拠に基づいて監査する。

Claude Codeに永続メモリやスケジュール実行を加え、常時稼働の自動エージェントを構築する。

単純なパイプラインからマルチエージェントDAGまで、自動Claude Codeループを設計する。

Apply backend architecture, API design, and database optimization patterns for Node.js and Express.

Node.js、Express、Next.js APIのバックエンド設計とデータベース最適化を指針として提供する。

パフォーマンスの基準値を測定し、PR前後の回帰や技術スタックの比較を行う。

複数セッションのプロジェクトを、自己完結したステップ付きの構築計画に変換する。

既存の文章からブランドの文体プロファイルを作成し、各種コンテンツで再利用する。

デプロイ後にブラウザ自動化でビジュアルテストとUI操作の検証を自動実行する。

BunとNodeの比較、移行方針、Vercelでの利用可否について助言する。

デプロイやマージ後に公開URLを監視し、回帰が起きていないか確認する。

運送会社のポートフォリオ管理、運賃交渉、パフォーマンス評価を支援する。

Cisco IOSの設定レビュー、ACL配置、安全な変更ウィンドウの検証を行う。

Persist project memory across Claude Code sessions, auto-loading context and tracking git activity deterministically.

Orchestrate parallel coding agents in isolated worktrees, monitoring their progress and reading structured reports.

Trace user-facing button flows to find bugs where features work alone but cancel out or leave inconsistent state.

Apply ClickHouse query optimization and analytics patterns for high-performance analytical workloads.

Create CodeTour files with anchored, step-by-step walkthroughs for onboarding, architecture, and PR reviews.

Generate onboarding guides with architecture maps, entry points, conventions, and a starter CLAUDE.md file.

Enforce consistent naming, readability, immutability, and quality review conventions across projects.

Apply general coding standards and best practices for TypeScript, JavaScript, React, and Node.js development.

Build Compose Multiplatform UIs with state management, navigation, theming, and platform-specific patterns.

Install, update, and reconfigure ECC across Claude Code, Codex, and Kimi while respecting each harness's features.

Reorganize X and LinkedIn follow lists and draft channel-specific outreach messages for high-value connections.

Adapt one source asset into platform-native posts, threads, scripts, and content calendars across channels.

Cache expensive file processing results using SHA-256 content hashes with automatic invalidation and service isolation.

Audit Claude Code context window usage and get prioritized recommendations for cutting token waste.

Extract reusable skills from sessions via a legacy Stop hook; deprecated in favor of v2.

Build atomic instincts from session hooks, scoped per project to avoid cross-project contamination.

Run safe schema and data migrations with rollbacks and zero-downtime deploys across major ORMs.

Set up CI/CD pipelines, Docker containers, health checks, and rollback strategies for production.

Diseñar aplicaciones Django de producción con APIs DRF, buen uso del ORM, caché y señales.

Configurar entornos Docker Compose seguros y multiservicio para desarrollo local y orquestación.

Escribir pruebas E2E estables con Playwright, Page Object Model y ejecución en CI/CD.

Evaluar sesiones de Claude Code con un marco formal de evals orientado a desarrollo.

Construir interfaces React y Next.js con buena gestión de estado y rendimiento optimizado.

Escribir código Go idiomático, mantenible y eficiente siguiendo las convenciones del lenguaje.

Escribir pruebas Go con tablas, benchmarks, fuzzing y cobertura siguiendo TDD.

Modelar entidades JPA/Hibernate eficientes con consultas, transacciones, auditoría y paginación.

Escribir Kotlin idiomático con coroutines, null safety y DSL builders mantenibles.

Escribir pruebas Kotlin con Kotest, MockK, coroutines, propiedades y cobertura con Kover.

Estructurar aplicaciones Laravel con servicios, Eloquent, colas, eventos, caché y API resources.

Asegurar aplicaciones Laravel con autenticación, validación, CSRF, subidas, secretos y límites de velocidad.

Desarrollar en Laravel guiado por pruebas con PHPUnit o Pest, factories y fakes.

Verificar proyectos Laravel con linting, análisis estático, cobertura y escaneos de seguridad.

Configurar Next.js 16 con Turbopack, aprovechando bundling incremental y caché en disco.

Optimizar consultas, esquemas e índices de PostgreSQL aplicando seguridad y buenas prácticas de Supabase.

Escribir Python idiomático y tipado según PEP 8, con type hints y código mantenible.

Escribir pruebas Python con pytest, fixtures, mocks, parametrización y cobertura exigida.

Construir servicios Quarkus 3.x con Camel, REST, CDI, Panache y procesamiento asíncrono.

Proteger aplicaciones Quarkus con OIDC, JWT, RBAC, validación, CSRF y gestión de secretos.

Test-drive Quarkus 3.x features and bug fixes using JUnit 5, Mockito, and REST Assured.

Run Quarkus build, static analysis, coverage, security scans, and native checks before release or PR.

Apply idiomatic Rust ownership, error handling, traits, and concurrency patterns to write safe code.

Write Rust unit, integration, async, property-based, and mocked tests with coverage, TDD-style.

Check authentication, user input, secrets, API endpoints, and payment code against a security checklist.

Structure Spring Boot backends with layered services, REST API design, caching, async work, and logging.

Harden Spring Security with auth, validation, CSRF, secrets, headers, rate limiting, and dependency checks.

Test-drive Spring Boot features with JUnit 5, Mockito, MockMvc, Testcontainers, and JaCoCo coverage.

Run Spring Boot build, static analysis, coverage, security scans, and diff review before release or PR.

Enforce test-driven development with 80%+ coverage across unit, integration, and E2E tests.

Run a complete verification system for Claude Code sessions to check work before finishing.

Recommend matching Claude Academy courses, tutorials, and use cases when users ask how to use Claude.

Create original seeded p5.js generative art with interactive parameter exploration for flow fields and particles.

Apply Anthropic's brand colors and typography to artifacts that need its look-and-feel or company design standards.

Create original posters, designs, and static visual art as PNG or PDF documents.

Reference Claude API and SDK details such as model IDs, pricing, streaming, tool use, and caching.

Prompt a discernment check after substantive advice or drafts so users verify claims before acting on them.

Guide structured co-authoring of proposals, specs, and decision docs through iterative refinement.

Create, read, edit, and format Word documents and templates, including tables of contents.

Guide distinctive frontend visual design choices in typography and aesthetics to avoid templated-looking UI.

Write status reports, leadership updates, and newsletters in the company's preferred formats.

Build MCP servers in Python or TypeScript that expose external APIs as well-designed LLM tools.

Read, merge, split, watermark, create, and fill PDF files, including forms and encryption.

Create, read, and edit PowerPoint decks and templates, extracting text or building presentations.

Create, evaluate, benchmark, and optimize skills, including tuning their descriptions for triggering.

Make animated GIFs for Slack within size limits, with validation tools and animation concepts.

Provide a blank starting point for authoring a new SKILL.md with its description and instructions.

Apply one of ten preset themes, or generate a new one, to slides, docs, or landing pages.

Build multi-component React, Tailwind, and shadcn/ui artifacts that need state management or routing.

Test local web apps with Playwright, checking UI behavior, capturing screenshots, and reading browser logs.

Open, edit, format, chart, and clean spreadsheets and CSVs, including computing formulas.

Apply lazy senior-dev mode: smallest complete change and a clear reply for any coding task.

Audit a whole repo for bugs, security holes, load failures, untested risk, and deletable code.

List every shortcut and ponytail comment as a debt ledger without changing any code.

Display measured code, cost, and speed savings from ponytail's benchmark in a one-shot report.

Show a quick reference of ponytail levels, skills, and commands for immediate lookup.

Review a change for logic, safety, load behavior, test coverage, speed, and unnecessary lines in plain English.

Designs stable APIs, module boundaries, and type contracts between frontend and backend.

Tests and debugs browser apps via Chrome DevTools, inspecting DOM, console, network, and performance.

Sets up CI/CD pipelines with quality gates, CI test runners, and deployment strategies.

Reviews diffs and pull requests across multiple quality dimensions before merging changes.

Refactor working code for clarity and maintainability without changing its behavior.

Writes a quality bar into CONSTRAINTS.md and flags agent changes that quietly weaken it.

Configures agent context and rules files to keep output quality high across sessions.

Finds root causes of failing tests, broken builds, and unexpected errors systematically.

Removes old systems and migrates users or database schemas without downtime, safely.

Records architecture decisions and API changes as ADRs for future engineers and agents.

Cross-examines decisions with fresh-context adversarial review to surface hidden failure modes.

Builds accessible, responsive, production-quality user interfaces with proper state management.

Structures commits, branches, and PRs, splitting messy working trees into atomic commits.

Turns vague ideas into sharp concepts through divergent and convergent thinking exercises.

Delivers multi-file changes in thin, verifiable slices, optionally behind feature flags.

Interviews the user one question at a time to uncover what they actually want.

Adds logging, metrics, tracing, and alerts so production behavior is diagnosable.

Profiles and fixes bottlenecks across frontend, backend, queries, and Core Web Vitals.

Breaks specs into ordered, implementable tasks with scope estimates and parallel work.

Audits input handling, auth, and data storage against OWASP Top Ten vulnerabilities.

Runs pre-launch checklists, monitoring setup, staged rollouts, and rollback strategies.

Verifies implementation choices against official documentation to produce authoritative, source-cited code avoiding outdated patterns.

Drafts specs, PRDs, and requirements with objectives and scope before any coding begins.

Drives code changes through red-green-refactor tests to prove behavior and catch regressions early.

Discovers and selects the right agent skill for a session or task at hand.

Maintains a cited local company wiki through the LLM Wiki plugin's tools.

Drive a real browser to take screenshots, read console and network output, and verify pages interactively.

Read, send, and reply to email through an assigned AgentMail inbox and confirm delivery status.

Submit an unstructured complaint about agent tooling proactively, without asking permission or previewing.

Plan a weekly editorial calendar mapping company goals to topics, owners, statuses, and verification notes.

Hands user-requested work to an existing Paperclip agent with a durable task reference.

Hands user-requested work to an existing Paperclip agent with a durable task reference.

Hands user-requested work to an existing Paperclip agent with a durable task reference.

Critique product design for job clarity, hierarchy, accessibility, and consistency, ranked by what to change first.

Updates project docs for recent code changes, adding release notes without rewriting unchanged sections.

Guide a user's first Paperclip task, proposing a plan and waiting for approval before hiring or executing.

Finds earlier Paperclip deliverables and attaches user feedback with human attribution.

Finds earlier Paperclip deliverables and attaches user feedback with human attribution.

Finds earlier Paperclip deliverables and attaches user feedback with human attribution.

Prepare a GitHub pull request with clean commits, clear title and body, verification notes, and review replies.

Rebuilds the wiki index from the actual page tree and logs reconciled counts without editing bodies.

Decide one next action per stale, blocked, or stalled inbox issue: resume, reassign, unblock, escalate, or close.

Call the Paperclip control plane API to check assignments, update issues, post comments, and delegate work.

Call Paperclip control plane API to check assignments, update issues, post comments, and delegate.

Read Paperclip task context, deliver documents, report blockers, and follow governance during heartbeats.

Manage a Paperclip company through chat, covering onboarding, agents, approvals, costs, and work review.

Generate, implement, or review capsule visuals, identicons, and status art while validating brand usage.

Decide whether plan work needs delegation and link justified tasks to owners and real dependencies.

Hire new Paperclip agents by comparing configs, drafting prompts, and submitting governance-aware hire requests.

Turns Paperclip activity bundles into project standups, decision logs, and durable wiki pages.

Creates, updates, comments on, and lists Paperclip tasks from Hermes using scoped credentials.

Store, retrieve, and organize durable knowledge across sessions using a file-based PARA memory system.

Research and prepare MCP vendor integrations with cited findings, approved content PRs, and governed connector PRs.

Write QA acceptance criteria and manual validation plans with explicit pass/fail evidence for a feature.

Follow Ramp's agent playbooks in Paperclip, with required human approval before any spend or account action.

Turn an agent's execution record into evidence-backed, minimal instruction or tool-description change proposals.

Write release announcements that lead with user impact and include clear upgrade and migration steps.

Summarizes a connected Paperclip team's agents, tasks, blockers, and pending approvals needing attention.

Summarizes a connected Paperclip team's agents, tasks, blockers, and pending approvals needing attention.

Summarizes a connected Paperclip team's agents, tasks, blockers, and pending approvals needing attention.

Write comments, plans, and documents in short ASD-STE100 sentences that readers understand on first pass.

Read Slack conversations and collaborate through the assigned Slack bot during tasks and routines.

Maintain agent status cards and compile interest prompts into bounded company search queries.

Quietly submit a concrete improvement suggestion after noticing repeated, generalizable friction in agent work.

Write a short plain-language status summary leading with concrete actions needed to unblock work.

Write or revise a plan with a clear outcome and verification steps before execution begins.

Ingests captured raw sources into the LLM Wiki as cited source, entity, concept, and synthesis pages.

Audits wiki for contradictions, orphans, broken links, and index drift, returning findings without auto-fixing.

Answer wiki questions with citations from pages and raw sources, flagging gaps and offering durable synthesis.

Create black-and-white low-fidelity UI wireframes as SVGs or viewer pages for quick layout review.

Build an artistic creative-studio website with classical imagery, cobalt skies, serif type, and cinematic scrolling.

Generate jingles, voiceovers, and sound effects by routing to music, speech, or SFX audio models.

Write long-form blog posts or case studies with masthead, figures, pull quotes, and byline.

Draft SOAP or narrative clinical case reports with plausible vitals, labs, and evidence-based plans.

Add a self-contained floating chat widget with welcome screen, social links, and message input.

Score any HTML design artifact across five dimensions and produce a radar-chart review report.

Build a single-file admin analytics dashboard with sidebar, KPI cards, and charts.

Build a dating dashboard with KPI headlines, match charts, and a community activity ticker.

Estimate intrinsic value and fair price targets for public companies using discounted cash flow analysis.

Create a two-page digital e-guide preview with cover, table of contents, lesson body, and step list.

Build a documentation page with side navigation, a scrollable article body, and a table of contents.

Create a branded product-launch HTML email with hero, CTA, and specs grid.

Produce an engineering runbook with alerts, dashboards, procedures, on-call rotation, and incident checklist.

Generate a quarterly financial report with KPIs, revenue and burn charts, and P&L summary.

Build a team-management dashboard with tabs, member table, charts, and light and dark themes.

Prototype a three-frame gamified mobile app with quest tiles, XP ribbons, and tab bar.

Build a GitHub repo analytics dashboard covering stars, contributors, issues, and pull requests.

Plan a new hire's first week on one page with schedule, buddy, learning track, and checklist.

Turn consulting diagnosis into a client-adoptable action plan with named owners committed to the roadmap.

Create a 30-day hospitality onboarding training deck with behaviors, practice, checks, and manager follow-up.

Draft a plugin marketplace business case covering user pain, options, tradeoffs, and success metrics.

Build a decision deck on bring-your-own-model choices covering cost, quality, and routing tradeoffs.

Create an incident retro deck covering the daemon-restart data bug, root cause, fix, and follow-ups.

Build an enterprise AI-adoption brief deck with risk controls, ROI, and rollout plan.

Create a demo-day fundraising deck with hook, traction, moat, and raise for investors.

Assemble a live-demo deck walking from a one-line prompt to a runnable design.

Draft a B2B sales deck proposing a launch and adoption plan for switching from closed cloud tools.

Generate a 16:9 HTML deck in CRT-terminal telemetry style with scanlines and hazard-red accents.

Generate a 16:9 HTML deck in warm editorial-minimalist style with serif type and hairline rules.

Creates an engineering deep-dive talk deck explaining agent streams, sandboxes, and artifacts for developers.

Build a hospital data-governance board briefing covering patient-data risk, controls, and required approval.

Builds a weekly growth review deck with stars, installs, activation metrics and recommended next moves.

Drafts a personal-manifesto story deck on a year of saying yes, for talk audiences.

Assembles a staff-engineer promotion packet covering scope, proof moments, artifacts, and impact.

Narrates a retro-arcade collecting journey deck covering the obsession, the hunt, and meaning.

Designs a curatorial biennale deck covering thesis, rooms, works, and visitor journey.

Rebuilds a messy startup pitch into a board-grade deck with page grammar and proof pages.

Produces an executive QBR deck on progress, stalls, and resource reallocation asks.

Presents a Series A growth deck showing traction curve, expansion motion, and venture-scale potential.

Writes a product-launch announcement deck with headline, proof points, and call to action.

Structures a product manager's year-end self-review with evidence-backed outcomes, learning, and asks.

Presents an economics thesis defense on minimum-wage employment effects, identification, and limitations.

Builds a renewal and seat-expansion business case with realized value, usage proof, and ROI.

Plans a community-growth campaign deck covering loops, content calendar, and pipeline math.

Documents a brand visual-identity system with logo, color, type, and consistency rules.

Scaffolds a customer onboarding workshop deck teaching first-value paths and retention habits.

Builds a magazine type-and-color system deck with grid, tri-tone palette, and layout kit.

Drafts a municipal tree-canopy policy briefing with public need, evidence, options, and funding decision.

Models OpenDesign unit economics and BYOK costs with assumptions, sensitivity, and scaling rationale.

Structures a grocery margin-recovery deck with governing thought, driver tree, priorities, and roadmap.

Draft a grant-ready CRISPR base-editing deck covering hypothesis, approach, milestones, and risks.

Build a three-case-study designer portfolio deck that shows craft and judgment for senior interviews.

Create a transit funding briefing for city council covering ridership need, plan, risk controls, and ask.

Structure a pollinator-decline capstone defense covering survey design, data, contribution, and caveats.

Assemble a wedding-anniversary photo essay with decade scenes, turning points, and a closing reflection.

Deliver a consultative-selling training with role-play flows and a daily habit that lifts conversion.

Present a brutalist festival poster-series case study showing concept, system, and multi-format scaling.

Teach staff to spot phishing tells, run a drill, and act within the first sixty seconds.

Compile a neighborhood zine on closing corner shops with portraits, local voices, and community impact.

Shape a Kyoto cherry-blossom photo essay tracing arrival, peak bloom, and the trip's defining moment.

Defend a civic transit-hub wayfinding project, covering its brief, design process, and final outcome.

Weigh options and risks of monetizing the plugin registry now versus holding, with a clear recommendation.

Sequence digital-transformation bets and an operating rhythm to guide a legacy insurer's modernization roadmap.

Document safety findings, the evidence chain, and corrective mandates for a regulator and plant leadership.

Present a photography studio's signature work, process, and packages to win prospective clients.

Argue how linear perspective reshaped early cartography, grounding each claim in sources and evidence.

Generate HTML video compositions with captions, narration, audio-reactive visuals, and scene transitions.

Assemble an investor pitch book with market map, moat, unit economics, and the funding ask.

Generate single-image posters, key art, and editorial illustrations saved as PNG or JPEG files.

Generates a printable invoice page with billing parties, line items, tax, and totals.

Builds a decision-grade lab-meeting deck on microbiome-sleep results, caveats, and the next experiment.