docs-anthropic
The docs-anthropic skill fetches and delivers official Claude Code documentation by matching user queries to the most relevant page from Anthropic's docs site. Use it when seeking guidance on skills authoring, lifecycle hooks, built-in tools, plugins, MCP channels, custom commands, or the interactive REPL environment. The skill searches a curated index of seven core topics and can fall back to a master index for discovering additional documentation pages.
git clone --depth 1 https://github.com/syahiidkamil/Software-Engineer-AI-Agent-Atlas /tmp/docs-anthropic && cp -r /tmp/docs-anthropic/.claude/skills/docs-anthropic ~/.claude/skills/docs-anthropicSKILL.md
# Anthropic / Claude Code documentation index The curated set of official Claude Code docs to reach for, with what each one covers. All live under `https://code.claude.com/docs/en/`. | Topic | URL | What it covers | | :-- | :-- | :-- | | **skills** | https://code.claude.com/docs/en/skills | Authoring Agent Skills: `SKILL.md` format and frontmatter, progressive disclosure (description always in context, body on invoke), `disable-model-invocation` / `user-invocable`, supporting files, `context: fork`, packaging. | | **hooks** | https://code.claude.com/docs/en/hooks | Lifecycle hooks — `PreToolUse`, `PostToolUse`, `UserPromptSubmit`, `PreCompact`, `SessionStart`, `Stop`, `SubagentStop`, `Notification` — their JSON stdin/stdout contract, exit codes, and `settings.json` wiring. **(`PreCompact` = the compact-hook event.)** | | **tools-reference** | https://code.claude.com/docs/en/tools-reference | Reference for the built-in tools (Bash, Read, Edit, Write, Glob, Grep, WebFetch, …) and how permissions gate them. | | **plugins-reference** | https://code.claude.com/docs/en/plugins-reference | Plugin structure: `plugin.json`, bundling commands / agents / skills / hooks / MCP servers, marketplace layout, skills-dir plugins. | | **channels-reference** | https://code.claude.com/docs/en/channels-reference | Building an MCP **channel** server that pushes webhooks, alerts, and chat into a session: capability declaration, notification events, reply tools, sender gating, permission relay. (Research preview.) | | **commands** | https://code.claude.com/docs/en/commands | Slash commands — built-ins plus custom commands in `.claude/commands/` (now merged into skills), frontmatter, `$ARGUMENTS`/`$N`, and `:`-namespacing (plugin-only). | | **interactive-mode** | https://code.claude.com/docs/en/interactive-mode | The interactive REPL: keyboard shortcuts, vim mode, multiline input, and session controls. | **Master index:** `https://code.claude.com/docs/en/llms.txt` — the full list of every docs page, for discovering anything not in the table above. ## How to use this skill Argument passed: **$ARGUMENTS** - **If a topic was given** (a keyword above, or something one of these docs would answer — e.g. `hooks`, `precompact`, `skill frontmatter`, `channels`, `slash command args`): pick the single best-matching URL, `WebFetch` it, and answer the question from the live page. Cite the URL. If nothing matches the seven, fetch the master `llms.txt` index, find the closest page, then fetch it. - **If the argument is blank:** present the table above as the index — no fetching. Offer to pull any specific page on request. Keep the verdict-open, no-overclaim floor: report what the docs actually say, and if a page has moved or 404s, say so plainly rather than reconstructing it from memory.
Activate ATLAS pattern recognition mode. Identify repeated patterns, extract essential characteristics, and create reusable abstractions from concrete examples.
Adversarially review something just built — presume it is broken and find where. Use after implementing a feature, finishing a build, or before shipping, or whenever ATLAS or Boss wants a hostile second opinion on a diff, a running app, or a whole codebase. Spawns clean-context reviewers with opposing lenses (correctness, security, empty-world) and reproduces every finding by running the target before believing it. Distinct from /code-review, which statically reads a diff and never runs the app.
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Stop and consult this skill whenever your response would include specific facts about Anthropic's products. Covers: Claude Code (how to install, Node.js requirements, platform/OS support, MCP server integration, configuration), Claude API (function calling/tool use, batch processing, SDK usage, rate limits, pricing, models, streaming), and Claude.ai (Pro vs Team vs Enterprise plans, feature limits). Trigger this even for coding tasks that use the Anthropic SDK, content creation mentioning Claude capabilities or pricing, or LLM provider comparisons. Any time you would otherwise rely on memory for Anthropic product details, verify here instead — your training data may be outdated or wrong.
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
Interview Boss about the project, then reason from first principles to design the ideal ATLAS operating identity/system-prompt for it — free to drop KISS/YAGNI/DRY/clean-architecture entirely when the project (and the LLM's own distribution) calls for a different mindset
Designs feature architectures by analyzing existing codebase patterns and conventions, then providing comprehensive implementation blueprints with specific files to create/modify, component designs, data flows, and build sequences
Deeply analyzes existing codebase features by tracing execution paths, mapping architecture layers, understanding patterns and abstractions, and documenting dependencies to inform new development