Project work can lose decisions, logs, and document state across agents. This skill guides Scribe MCP setup, documentation updates, and progress logging.

Repository conventions are often scattered across guides, formatter settings, linter rules, hooks, and CI files. This skill consolidates those sources into one evidence-based standards summary.

Specification requirements are difficult to locate and trace across repositories. This skill extracts requirement IDs, titles, paths, links, and hashes into a sorted manifest.

Teams lose time when models, APIs, and documentation use inconsistent business terms. This skill gives Claude, Codex, and Claude Code a shared domain glossary and relationship map.

Code changes often leave README files, changelogs, and docstrings outdated. This skill reviews implementation changes and updates affected documentation using repository conventions.

Documentation sites expose content through different platforms and formats. This skill identifies the source type and recommends a practical fetch and registry strategy.

Finding relevant local documentation can waste time and context. This skill guides targeted retrieval, source consolidation, and web fallback when local coverage is insufficient.

Large documentation sets waste context when assistants load broad files or guess paths. This skill follows hierarchical indexes to find targeted local pages before web search.

Architecture discussions become confusing when diagrams mix detail levels or omit relationships. This skill creates structured C4 views and reviews them against consistent guidelines.

JavaScript-rendered documentation can hide links from static fetches. This skill selects available browser tools and extracts structured navigation after pages render.

Teams need clear technical diagrams that show systems, networks, and cloud services. This skill guides AI tools to produce dark standalone HTML diagrams with inline SVG.

Project structure can be hard to follow during AI-assisted work. This skill guides Claude, Codex, and Claude Code to read architecture, conventions, patterns, and gotchas.

Teams often repeat design debates or miss prior rationale. This skill guides Claude, Codex, and Claude Code to review decision records before new choices.

Architecture decisions can become unclear when requirements, team skills, and future growth compete. This skill compares suitable patterns and helps document a clear, reasoned decision.

Teams lose trust when blockers are vague or hidden. This skill turns failures, fixes, and health metrics into clear reports.

Documentation drifts when code, APIs, and agents change. This skill guides Claude, Codex, and Claude Code through focused updates for README, API, architecture, and changelog files.

Creating a skill can be difficult when its files, metadata, and references are unclear. This guide explains a practical structure for authoring organized Claude Code skills.

Architecture knowledge often lives only in code and loses context during refactors. This skill creates concise architecture docs, diagrams, and component references for Claude, Codex, and Claude Code workflows.

Development sessions are hard to review when details stay in chat history or local logs. This skill turns Chronicle data into structured Obsidian notes with summaries, metadata, and links.

Agent systems can become unreliable when prompts, tools, history, and retrieval compete for context. This skill explains context components and practical patterns for budgeting, progressive disclosure, and validation.