发现 Skills

Analyze images with AI - extract text, describe content, detect objects

Use when co-authoring documentation with users through a structured workflow. Triggers include requests to write documentation, proposals, technical specs, decision docs, RFCs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers.

Use when executing implementation plans with independent tasks in the current session

Create a new App Store Connect app record via browser automation. Use when there is no public API for app creation and you need an agent to drive the New App form.

Surface concrete in-task-collaboration protocols when the user describes an AI workflow informally — URL-as-entity-reference, durable PR-comment threads as session memory, "fit the protocol" basics. Trigger when the user names entities by colloquial label instead of stable URL, asks "how should I structure this for Claude", pastes a screenshot when a URL would do, or describes a multi-step Claude workflow without a durable handle. Apply tactics reactively, not as a checklist.

Delete tests that don't catch real bugs — the inverse of TDD. Use when reviewing legacy test suites, slow CI investigations, refactor-driven test sweeps, or evaluating whether a test the type system already covers should stay. Thesis — a test exists only if removing it would let a real bug reach production.

Audit memory directory for structural issues (orphans, dangling refs, duplicates, missing sections, oversized entries) and staleness against session-history transcripts; report-first, fix-on-confirmation. Use when the user says "audit memory", "memory hygiene", or "find stale/duplicate memories".

Synthesize current conversation context and codebase understanding into a PRD artifact (markdown file by default, GitHub issue via flag). Trigger when the user asks for a PRD, requirements doc, feature spec, or wants to crystallize an in-flight discussion into a durable artifact before planning. PRD precedes implementation planning.

Use when automating browser workflows — form filling, file upload/download, login session reuse, multi-step flows

Use when extracting data from JavaScript-rendered pages, handling pagination, or scraping dynamic content

Performs API inventory and discovery to identify all API endpoints in an organization's environment including documented, undocumented, shadow, zombie, and deprecated APIs. The tester uses passive traffic analysis, active scanning, DNS enumeration, JavaScript analysis, and cloud resource inventory to build a comprehensive API catalog. Maps to OWASP API9:2023 Improper Inventory Management. Activates for requests involving API discovery, shadow API detection, API inventory audit, or attack surface mapping.

Foundational patterns for building accessible autocomplete/combobox components with state management, ARIA patterns, keyboard navigation, async suggestions, and framework-agnostic implementation. Use when building autocomplete inputs, command palettes, search inputs, select replacements, or any dropdown with keyboard navigation and suggestions.

AI-native planning framework for autonomous project decomposition. Accepts free text to decompose into epics, features, stories, and tasks. Subcommands: pending, loop, plan. Without arguments shows decision tree. Tasks are self-contained units with technical specs and binary acceptance criteria. This skill should be used when the user says "descomponer en features", "crear roadmap de X", "estructura de X", "planificar implementación de X", "qué sigue", "ver roadmap", "ver progreso", "qué falta", "tasks pendientes", "loop de tasks", "ejecutar pendientes", "implementar tasks", "roadmap loop", "ejecutar roadmap", "crear roadmap del plan", "materializar plan", or provides free text describing work to decompose.

A complete skill for E2E testing

Comprehensive code review workflow - parallel specialized reviews → synthesis

Problem-solving strategies for rings in abstract algebra

Analyzes code for bugs, security vulnerabilities, performance issues, and style violations. Use when the user asks for a code review, feedback on specific files, optimization suggestions, or a security audit.

Diagnose common nvm issues on macOS, including profile loading, PATH priority, and permissions.

Install the Specify CLI on the host machine (uv tool install or uvx one-time); supports multiple OSes, persistent or one-time install, and corporate or restricted-network environments. Use when the user says "install Spec Kit", "install Specify CLI", or "specify command not found".

Implements and analyzes Rust async runtime patterns including tokio, async-std, and custom executors for high-performance concurrent systems.

"NATS in Cloud Native Messaging - cloud native architecture, patterns" pitfalls, and best practices

Software R&D pipeline orchestrator that walks a knowledge graph / wiki, an intent-driven OpenSpec change, Hermes Kanban execution, and ingest synthesis as a single closed loop. Use when starting, continuing, or inspecting a development effort whose source-of-truth is a wiki and whose execution is a durable kanban board. Do not use for one-off code edits, dependency bumps, or non-spec work.

Author Gherkin-style executable specifications for a scientia change, one spec.md per capability. Reads the manifest core (especially ubiquitous-language and tradeoffs slices), enforces gherkin-authoring discipline (single When per scenario, observable Then, named personas, Markdown wrappers preserved), and writes openspec/changes/<tenant>-<change-id>/specs/<capability>/spec.md. Use after proposal.md is merged to trunk and before scientia-intent-design.

RDKit molecular manipulation and visualization within Coot's Python environment. Use when working with Coot and need to (1) Create RDKit molecules from Coot monomers, (2) Modify molecular structures (e.g., atom substitution), (3) Generate 2D chemical structure diagrams, (4) Perform cheminformatics operations on ligands or small molecules loaded in Coot.