Railway log access and analysis for debugging and monitoring. Covers build logs, deploy logs, runtime logs, HTTP logs, filtering, search, and external export. Use when viewing logs, debugging Railway deployments, investigating errors, analyzing HTTP requests, filtering log output, or exporting logs to external systems.

Complete design system guide for Dr. Sophia AI's botanical brand identity. Includes official typography (Livvic/Onest), color palette (#00B368 green spectrum), shadow systems, gradients, and component templates. Use when creating or modifying UI components, implementing brand colors, or ensuring design consistency across the application.

Use when user wants single voice input capture without spoken response - invokes voice listening, transcribes speech, responds in text only

Search, create, and manage notes in the Obsidian vault with wikilinks and index notes. Use when user wants to find, create, or organize notes in Obsidian.

Orient users to the i4h workflow runtime and route them to the correct stage skill. Use for architecture, support, or where-to-start questions; do not execute a known stage.

Run the maintained workflow data-to-policy pipeline from recording through checkpoint validation. Use for full end-to-end requests; do not use for one individual stage.

Serve and visually inspect a converted LeRobot dataset in the browser. Use for videos and state/action timelines; do not use for raw workflow HDF5 or incomplete conversion output.

Safely integrate private local knowledge systems using explicit data classification, sanitized fixtures, compatibility checks, and reversible writes. Use before indexing, syncing, retrieving from, migrating, exporting, or changing a personal vault, memory runtime, note template, or knowledge connector.

Transforms a research artifact or feature topic into a compact implementation plan with task breakdown, dependency mapping, test specifications, and acceptance criteria. Use when planning a non-trivial feature, designing architecture, writing a technical spec, or breaking down implementation steps before coding. Produces a phased plan with Agent Context blocks and a beads task graph for incremental execution.

Post-session reflection skill. Reads recent git history, artifacts, and context files to extract learnings and produce improvement proposals for skills, CLAUDE.md, hooks, and plan templates. Use after any substantive session.

Sub-architect Backend. Invocado por /architect orchestrator. Lee 01-spec.md + (02-design-ui.md si mixed) + story YAML. Produce 03-arch-be.md con: endpoints, DTOs Pydantic, SQLA models, migrations idempotent, services, repositories, tests requeridos, cross-cutting (tenant isolation, currency, master-data, PII). Cross-module audit obligatorio. Activa cuando /architect spawna o user dice: '/architect-be', 'arq backend', 'diseña BE'.

Architect orchestrator Nicolify. Lee 01-spec.md + 02-design-{ui,agentic}.md. Decide qué surfaces toca (BE/FE/agentic). Spawna sub-architects en paralelo (/architect-be, /architect-fe, /architect-agentic). Reúne sus 03-arch-*.md y produce 04-tickets.yaml ordenado con dependencias + owner_eligibility (qwen vs Opus) + acceptance criteria + quality gates. Activa cuando user dice: '/architect', 'diseñemos la arq', 'tickets', 'qué tickets salen', 'arquitectura técnica', 'cómo lo construimos técnicamente'.

Generate, audit, and compress AI-optimized documentation for codebases. This skill applies research-backed principles from three peer-reviewed papers to create documentation that maximizes AI agent performance instead of degrading it. Use this skill whenever the user asks to "write docs", "create a CLAUDE.md", "document a module", "audit documentation", "compress docs", "optimize docs for AI", "write module docs", "create domain documentation", "review doc quality", or mentions documentation bloat, token waste, or AI context efficiency. Also trigger when the user references "ai-docs principles", "telegram style docs", or wants to apply the "Lost in the Middle", "Less is More", or "AGENTS.md" research findings.

Use when a selected electronic parts list must be validated, enriched, exported, or made procurement-ready.

Review Web Audio API, Tone.js, DSP, and sound synthesis code for audio graph lifecycle leaks, glitch/pop artifacts, clipping prevention, parameter ramps, and browser AudioContext safety.

Implements a mainspec end-to-end by auto-detecting mode. Sequential mode (≤3 slices) commits slices in order on the current `feature/<feature>` branch. Parallel mode (>3 slices) uses dependency-aware tiered execution with per-slice worktrees, branches, PRs, and auto-merge into the feature branch. Agent-first — invoked headless by the harness dispatcher with the feature slug as its single argument. No human-in-the-loop, no approval gates.

Turn an open-ended idea into a PRD plus a runnable definition of done. Use when a developer wants to start a new feature, "file a PRD", "spec out" or "scope" an idea, think through what to build, or kick off the harness for a piece of work. Produces prds/<feature>/prd.md + an executable prds/<feature>/run-prd-test.sh on a prd/<author-slug>/<feature> branch. The one human-attentive skill in the chain.

Update the project's long-term memory after a merge to main. Reads the merged diff and the current memory, then writes Expert shards, discovered invariants, candidate lints, and AGENTS.md pointers — all on a reviewable learn/<sha> PR. Use post-merge (the harness invokes it automatically) or with --rebuild to regenerate memory from scratch. Triggers - learn, expert-update, update memory, update expert, post-merge memory, self-improve (project)

Generates requirements documentation (user stories, acceptance criteria, out-of-scope) for a feature in an AI-native MVP project. Invoke before any implementation of a new feature begins — a feature without requirements documentation should not be built. If the developer mentions wanting to add or implement something new, trigger this skill before any code is written.

Allows to interact with web pages by performing actions such as clicking buttons, filling out forms, and navigating links. It works by remote controlling Google Chrome or Chromium browsers using the Chrome DevTools Protocol (CDP). When Claude needs to browse the web, it can use this skill to do so.