Repository documentation can be difficult to inspect quickly during coding work. This skill routes GitHub repositories through GitMCP so Claude, Codex, or Claude Code can fetch docs, search docs, and search code.

Large development plans are slow when every task waits on one agent. This skill parses plan dependencies and coordinates parallel subagents to complete, validate, log, and commit work.

Manual browser testing and data extraction can slow agent workflows. This skill provides structured commands for navigation, interaction, snapshots, and session control.

Microsoft 365 tenant administration requires careful planning, repeatable scripts, and security review. This skill creates setup plans, checklists, and PowerShell drafts for tenant, user, license, and security workflows.

Long documents can be hard to summarize without losing meaning or biasing the result. This skill compares multiple model perspectives and records the reasoning trail.

Messy commit history slows review and makes rollback harder. This skill helps Claude, Codex, and Claude Code choose a commit strategy and write clear messages.

Teams need fast platform-ready content without rewriting every idea for each channel. This skill gives Claude, Codex, and Claude Code structured prompts and inference.sh workflows for social videos, images, captions, hooks, and thumbnails.

Claude Code hooks are powerful but easy to format incorrectly or secure poorly. This skill gives Codex and Claude clear guidance for hook files, events, validation, exit codes, and PRPM publishing.

Teams need repeatable prompts and settings for short AI video generation. This skill guides Claude, Codex, and Claude Code through Wan 2.7 requests using the RunComfy CLI.

Video teams often need to adapt footage to new aspect ratios without cropping important action. This skill guides RunComfy CLI and ComfyUI workflow choices for spatial video outpainting.

Teams often waste time choosing the wrong image-to-video model for a creative request. This skill routes each request to a suitable RunComfy model and supplies model-specific prompting guidance.

Python dependency workflows can be slow and inconsistent across local, CI, and Docker environments. This skill gives practical uv commands for faster installs, lockfiles, virtual environments, and migration paths.

Land and trespass evidence review requires consistent metadata, provenance, and factual language. This skill guides agents through a strict queue workflow for reserve reduction, survey error, and railway taking records.

Legal evidence review requires careful separation of facts, quotes, provenance, and conclusions. This skill guides Claude, Codex, and Claude Code through structured coercion and duress task review with strict citation discipline.

Large EPUB files are difficult to review, quote, or reorganize as a single document. This skill splits an EPUB into numbered markdown chapter files for easier editing and analysis.

PDF layout is hard to inspect from extracted text alone. This skill creates readable ASCII previews that combine page images and extraction metadata.

Manual browser work is slow and hard to repeat across pages, forms, and test flows. This skill gives Claude, Codex, and Claude Code structured Playwright MCP commands for controlled browser automation.

Teams need reliable patterns for adding AI features to Next.js apps. This skill provides focused guidance for chat, structured output, generative UI, and background agents using Vercel AI SDK.

PowerPoint editing often needs precise layout control, template awareness, and file validation. This skill provides workflows for creating, modifying, inspecting, and checking PPTX files with HTML, OOXML, and helper scripts.

Docusaurus deployments can fail because configuration, build output, and GitHub Pages settings must align. This skill guides local validation, workflow setup, and deployment checks before publishing.