Latest Agent Skills

Audits websites for technical, on-page, and content SEO issues, providing prioritized recommendations to improve organic search performance.

Ingests video and audio from files, URLs, live feeds, or desktop capture; indexes and searches moments with timestamps; transcodes, edits timelines, generates media assets, and emits real-time alerts.

Manage Uncloud clusters with the `uc` CLI: bootstrap machines, deploy Compose services, publish ports, configure Caddy ingress, scale, inspect logs, and manage volumes.

Records polished UI demo videos of web applications using Playwright, with an injected cursor overlay, natural pacing, and a discover-rehearse-record workflow.

Convenes four independent advisor voices to surface structured disagreement on ambiguous decisions, tradeoffs, and go/no-go calls before choosing a path.

Analyzes marketing pages and provides actionable conversion rate optimization recommendations, covering value proposition, headlines, CTAs, trust signals, and friction points.

Builds elaborate, multi-component claude.ai HTML artifacts using React, Tailwind CSS, and shadcn/ui, with scripts to initialize a project and bundle it into a single HTML file.

Integrates with the X (Twitter) API to post tweets and threads, read timelines, search content, and track analytics using OAuth authentication and rate-limit handling.

Provides a creative-direction layer for music videos and short-form edits, defining an angelcore/cloud-trance/hyperpop aesthetic with mood, color, light, and beat-synced editing grammar, and orchestrating video production skills into a pipeline.

Apply David Ogilvy's advertising principles to write or review persuasive copy, covering positioning, promises, headlines, body copy, and visual logic.

Audits and improves search visibility through technical SEO, on-page optimization, structured data, Core Web Vitals, and content strategy.

Persistent per-project memory for Claude Code. Auto-loads project context on session start, tracks sessions with git activity, and writes to native memory. Commands run deterministic Node.js scripts — behavior is consistent across model versions.

Creates animated GIFs optimized for Slack with constraints, validation, and animation techniques using PIL.

Discovers, audits, repairs, adapts, and designs bounded AI-agent loops with explicit triggers, actions, verification, stopping conditions, guardrails, and handoffs.

Applies Anthropic's official brand colors and typography to artifacts such as presentations, ensuring a consistent look and feel.

Guides distinctive, intentional visual design for new or existing UI, covering aesthetic direction, typography, and avoiding templated defaults.

Guides users through a structured three-stage workflow for co-authoring documents, from context gathering to reader testing.

Creates algorithmic art with p5.js, starting from a generative philosophy and expressing it through seeded randomness, noise fields, and particle systems.

Interact with and test local web applications using Playwright, including verifying frontend functionality, debugging UI behavior, capturing screenshots, and viewing browser logs.

Writes internal communications such as status reports, leadership updates, 3P updates, company newsletters, FAQs, and incident reports using company-preferred formats.

Applies curated color palettes and font pairings to presentation slide decks and other artifacts, with ten pre-set themes and the ability to generate custom themes on the fly.

Guides users through creating, editing, and optimizing agent skills, including drafting, testing, evaluating, and improving skill descriptions for better triggering.

Creates original visual art in .png and .pdf documents by first generating a design philosophy and then expressing it visually, with minimal text and expert craftsmanship.

Guides the creation of high-quality MCP servers, covering design, implementation, testing, and evaluation for Python and TypeScript.

Reference for building LLM-powered applications with the Claude API and Anthropic SDKs, covering model selection, parameters, streaming, tool use, and migration.

Creates, edits, and analyzes spreadsheet files (xlsx, xlsm, csv, tsv) with formulas, formatting, and data cleaning, including recalculation and error checking.

Creates, edits, and analyzes PowerPoint .pptx and .potx files, including generating decks with pptxgenjs, editing slide XML, and validating output.

Creates, edits, and analyzes Word documents (.docx/.dotx) using docx-js, XML manipulation, and conversion tools, including tracked changes and comments.

Reads, extracts, merges, splits, rotates, watermarks, creates, encrypts, and OCRs PDF files using Python libraries and command-line tools.

Checks project progress, summarizes recent work and what's ahead, then routes to the next action by executing an existing plan or creating a new one.

Captures ideas, tasks, or issues during a GSD session as structured todos for later work, enabling a thought-to-capture-to-continue flow without losing context.

Dispatches a code-reviewer subagent to evaluate work against requirements and plan, then acts on the feedback before merging.

Builds a scheduled, AI-powered data collection agent that scrapes public sources, enriches results with Gemini Flash, and stores them in Notion, Sheets, or Supabase.

Builds reliable browser automation skills by iteratively running a browsing task, reading the trace, and improving the navigation strategy until it passes consistently.

Manages a milestone from a single terminal, showing a phase dashboard and dispatching discuss, plan, and execute actions.

Guides developers through Socratic exploration of ideas, offering research and routing outputs to planning artifacts.

Archives phase directories from completed milestones into versioned archive folders, showing a dry-run summary and moving directories only after confirmation.

Ingest, index, search, edit, and generate video and audio assets from files, URLs, RTSP feeds, or desktop capture, with real-time alerts and stream links.

Records polished demo videos of web applications using Playwright, with an injected cursor overlay, natural pacing, and a discover-rehearse-record workflow.

Convenes a four-voice council of subagents to surface structured disagreement for ambiguous decisions, tradeoffs, and go/no-go calls.

Guides technical evaluation of code review feedback, emphasizing verification before implementation and reasoned pushback over performative agreement.

Repairs failed task verification by choosing a strategy: retry with adjustments, decompose into sub-tasks, prune infeasible tasks, or escalate to the user.

Deploys serverless browser automation as cloud functions using Browserbase, including project setup, local testing, and publishing via the bb CLI.

Translates visa application documents from images into professional English PDFs, handling image conversion, rotation, OCR, and translation.

Checks for GSD updates via npm, shows the changelog between installed and latest versions, and performs a clean reinstall after user confirmation.

Runs independent AI CLI reviewers over phase plans and merges their feedback into a REVIEWS.md file.

Imports external plan files into a GSD-structured project, detecting conflicts against existing roadmap, requirements, and locked decisions before writing a validated PLAN.md.

Validates the integrity of a project's .planning/ directory, reporting missing files, invalid configurations, and inconsistent state, with optional auto-repair of safe issues.

Adopt Swift 6.2's concurrency model with patterns for MainActor isolation, isolated conformances, and explicit background offloading to eliminate data-race errors.

Displays comprehensive project statistics including phases, plans, requirements, git metrics, and timeline.

Runs focused experiments to validate feasibility of an idea, saving artifacts to .planning/spikes/ and supporting both idea and frontier modes.

Executes small ad-hoc tasks with GSD guarantees, spawning planner and executor agents, tracking tasks in .planning/quick/, and updating STATE.md. Supports composable flags for discussion, research, plan-checking, and verification.

Triage and review open GitHub issues and PRs against project contribution templates, producing a compliance report and optionally labeling, commenting, or closing items.

Interact with X (Twitter) programmatically: post tweets and threads, read timelines, search, upload media, and track engagement using OAuth 1.0a or Bearer token authentication.

Runs adversarial UI tests in a real browser, covering functional, accessibility, responsive, and UX checks, with diff-driven, exploratory, and parallel workflows.

Automates web browser interactions via the browse CLI, including navigation, form filling, screenshots, and scraping, with support for local and remote Browserbase sessions.

Creates brand identities, logos, corporate identity programs, HTML presentations, banners, icons, and social media images using AI generation and HTML/CSS-to-screenshot workflows.

Rolls back Open-workspace phase or plan commits using a phase manifest with dependency checks and a confirmation gate, preserving history via git revert.

Creates a pull request from completed phase work, generates a rich PR body from planning artifacts, optionally runs code review, and prepares for merge.

Assesses a codebase by spawning a single mapper agent for a chosen focus area, producing targeted documents in .planning/codebase/.