PyTorch docstrings require precise reStructuredText, tensor notation, and project conventions. This skill guides consistent signatures, arguments, cross-references, formulas, warnings, and examples.

Plugin authors need a reliable structure before adding commands, agents, skills, hooks, or MCP servers. This skill explains the expected layout, manifest fields, naming rules, and portable path patterns.

Plugin authors need a safe way to store per-project settings without committing user-specific state. This skill explains a local Markdown and YAML frontmatter pattern for Claude, Codex, and Claude Code plugins.

Codex guidance becomes stale, duplicated, and difficult to trace. This skill maintains a review-first wiki with source provenance, explicit knowledge states, safe paths, and deterministic linting.

SOP documents are often difficult to teach and verify consistently. This skill turns supplied procedures into grounded, interactive PersonWise courses with reviewed content.

Support teams repeat the same explanations, while customers struggle to find reliable answers. This skill turns approved source materials into grounded, interactive PersonWise courses.

Compliance teams need accurate courses grounded in current policies. This skill converts supplied materials into interactive training with review checkpoints and controlled publication.

OpenAI products, models, and APIs change quickly, making remembered guidance unreliable. This skill retrieves current official documentation and provides concise, cited answers.

AI skills can fail when instructions are vague or untested. This skill guides test-driven skill writing with pressure scenarios and verification loops.

Teams need repeatable guidance to build useful AI skills without adding unnecessary context. This skill provides a structured process for planning, writing, validating, and packaging skills.

Unfamiliar repositories slow onboarding, reviews, and documentation work. This skill guides Claude, Codex, and Claude Code through structured architecture analysis.

AI writing can erase sources or omit consent boundaries when discussing culturally rooted knowledge. This skill preserves attribution, provenance, revocation, and human authority.

Plain PDF pipelines often produce generic layouts. MiniMax PDF creates, fills, and reformats PDFs with consistent design tokens and print-ready output.

Technical documentation often becomes formal, vague, or difficult to scan. This skill helps Claude, Codex, and Claude Code write clear Metabase documentation with focused steps.

Documentation PRs can drift from the Metabase style guide and create unclear guidance for readers. This skill checks changes for tone, structure, formatting, links, examples, and useful review feedback.

Teams lose clarity when one concept has several names or one term has several meanings. This skill creates a canonical DDD glossary from your conversation.

Planning work from a long conversation can lose decisions, constraints, and testing intent. This skill turns existing context into a structured spec and publishes it to the issue tracker.

Important product decisions often remain scattered across conversations and code context. This skill synthesizes them into a structured PRD and publishes it to your issue tracker.

Unresolved assumptions weaken plans and technical designs. This skill conducts a focused interview and uses domain modeling to produce ADRs and a glossary.

Inconsistent domain terms create unclear requirements and fragile boundaries. This skill builds a shared glossary, maps contexts, and records consequential decisions.