Skip to content

btseytlin/ultrapack / skills/udocument

udocument

Unverified

Use when writing or editing documentation — project docs, CLAUDE.md, READMEs, SKILL.md, docstrings, inline comments. Auto-triggers on `.md` files and docstring edits. Enforces lead-with-why, kill stale content, lists over tables, no aspirational sentences.

Document

Documentation explains how to use and maintain the system. Apply these rules to project docs, agent guidance, skills, docstrings, and comments.

Core rules

  • Lead with the problem or purpose, then describe behavior and usage.
  • Prefer one concrete example over repeated explanation.
  • Remove stale descriptions of behavior that no longer exists. Do not add historical notes to preserve them.
  • Document supported behavior, not aspirations.
  • Use lists instead of tables unless the user requests a table.
  • Write short, grammatical sentences. Remove framing and repeated instructions.
  • State concrete constraints. "Dataset must not import training" is clearer than "respect boundaries".
  • Consolidate duplicated guidance and link to its owner.

Project docs

Answer what the project does, why it exists, how to use it, and where related documentation lives. Include a minimal working example. Keep a README under 200 lines, moving detail into docs/ when needed.

Agent guidance

Put project-wide rules, constraints, style preferences, and pointers in CLAUDE.md or its equivalent. Per-task instructions belong in docs/tasks/<slug>.md.

Skills

  • Preserve name and description frontmatter. The description determines when an agent loads the skill.
  • Cover when to use it, the process, key rules, required output, and terminal state.
  • Use descriptive headings. Aim for a body under about 150 lines without removing required behavior.

Docstrings

For a large class or module, explain its responsibility and boundaries:

class DatasetBuilder:
    """Build dataset shards from raw episodes.

    Validate inputs, transform records, and checkpoint progress.
    Training-time loading belongs in training/loader.py.
    """

For a small utility, one line is enough. Omit docstrings that merely repeat an internal helper's name.

Inline comments

Explain non-obvious reasons: a workaround, required ordering, hidden invariant, or surprising choice. Remove narration such as # increment counter, change-history notes, and placeholder apologies.

Keep a concrete historical cause when it explains a current constraint. "Guard against overflow that corrupted billing totals" explains a guard. "Added in the billing cleanup" does not.

Process

  1. Identify the audience and the question the document must answer.
  2. Choose a list, example, or paragraph that fits the content.
  3. Draft concise instructions grounded in the implementation.
  4. Remove text that adds no instruction or explanation. Do not cut to a percentage target.
  5. Include a concrete example where it helps the reader act.
  6. Check references and remove stale, aspirational, or duplicated claims.
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 464kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 196kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 196kUpdated Claude CodeCodexVerifiedInstall