Explore skills

Transcribes and translates audio/video files locally using OpenAI Whisper.

Posts articles to DEV.to using AppleScript Chrome control, ideal for publishing technical blog posts, showdev articles, or open source project announcements.

Generates changelogs and release notes from commits and tags. Use when releasing a version, writing release notes, or maintaining a CHANGELOG.

Creates and updates READMEs, API documentation, runbooks, and in-code documentation for any project type or audience. Use when documenting a project, API, or process, or when documentation is requested.

Manage metadata of photo, video, and audio files with this skill, including adjusting capture timestamps, handling GPS geotags, and stripping private data.

The Instinct Engine scans recent sessions, journals, and decisions for recurring patterns, turning them into concrete captures like CLAUDE.md rules or skill improvements. It runs after weekly reviews, when patterns harden, or semi-autonomously via session-end triggers.

Enforces documentation-first development by identifying and updating impacted documentation alongside code, recording changes in CHANGELOG.md, and removing stale or contradictory docs before any code change. Apply whenever code is added, modified, or removed, including "quick fixes".

Record narrated tutorial videos using the Vorec Recorder macOS app. It's ideal for creating new tutorials, demos, screencasts, or walkthroughs, and for updating narration or translations on existing Vorec projects.

Senior embedded software engineer specializing in firmware and driver development for ARM Cortex-M microcontrollers (Teensy, STM32, nRF52, SAMD). Decades of experience writing reliable, optimized, and maintainable embedded code with deep expertise in memory barriers, DMA/cache coherency, interrupt-driven I/O, and peripheral drivers.

Write documentation following Metabase's conversational, clear, and user-focused style. Use when creating or editing documentation files (markdown, MDX, etc.).

Detects and removes unnatural patterns specific to AI-generated Japanese text, transforming it into a natural, human-like writing style. This Japanese adaptation of the humanizer corrects issues like excessive humble expressions, reliance on bullet points, formulaic endings, and common AI vocabulary.

Automatically adds user permissions to Feishu documents upon creation (e.g., multi-dimensional tables, documents, spreadsheets, folders, cloud files, knowledge base nodes) or when users report access issues. Applicable to OpenClaw Agent.