CodeWhale

Run cargo check on the current Rust project to find compile errors

Break a large, parallelizable goal into bounded work units, coordinate existing agent/worktree machinery, integrate, and verify. Explicit-only.

Generate a small set of independent candidate solutions in worktrees, judge them against one explicit rubric, and apply the winner only after PASS verification.

Help a new contributor get productive on this checkout - inspect sync state against main, build, run the repository's exact verification gate, and produce a local what's-new digest. Never fetches, pulls, or modifies a dirty tree on its own. Explicit-only.

Choose and produce an appropriate chart/dashboard/visual explanation from data, with legible encodings and source/assumption notes.

Reproduce, minimize, localize, identify root cause, and distinguish diagnosis from an authorized fix. Prefer root-cause over symptom patches.

Strategic delegation for multi-step coding, research, or verification work. Use when a task can be split into parent reasoning plus focused sub-agent execution through the agent tool.

Read release notes/changelogs, update a defined dependency scope, handle breaking changes, and verify. Explicit-only; no broad update-everything by inference.

Write or update repository/product documentation: README, user guides, API docs, architecture, migration notes, and changelog material. This is not the Word-document artifact skill.

Compatibility alias for the docx skill. Prefer loading docx.

Create/edit/inspect Word documents using available managed or host capabilities.

Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.

Use when managing, triaging, restarting, escalating, or summarizing Codewhale Agent Fleet runs and workers.

Produce intentional, responsive, accessible UI work and perform visual QA instead of generic component assembly.

Write a compact, decision-ready handoff so the next session (or the user) can continue without reconstructing the current one. Use when the session is ending, context is running low, the user asks for a handoff / "pass the baton" / "hand off", or a long-running operation needs a durable state checkpoint.

Route a "how do I use Codewhale" question to the installed help, config, and doctor surfaces instead of reciting a manual from memory. Explicit-only.

Carry an authorized, defined request or approved plan through scoped edits and proportionate verification. Skill loading is not write authorization by itself.

Ask one useful structured question at a time only when material product/implementation choices are genuinely missing; remember answers and produce a brief/spec. Discoverable facts should be investigated instead of asked.

Design, build, configure, or debug Model Context Protocol servers for codewhale, including stdio and HTTP/SSE transports.

Read, extract, split, merge, rotate, watermark, fill, OCR, or create PDF files with verification of page counts and text extraction.

Turn a sufficiently understood task into an ordered implementation plan with dependencies and verification. Orchestrate CodeWhale’s native plan state; do not build a parallel planner.

Scaffold a local Codewhale plugin bundle with a versioned manifest, namespaced Skills, and an explicit trust review.

Create/edit/inspect/verify slide decks and PPTX presentations.

Compatibility alias for the pptx skill. Prefer loading pptx.

Prepare a named version: preflight, version consistency, build/package, smoke test, checksums/notes, and release readiness. Publishing/tagging/deploy need separate authorization. Explicit-only.

Current external research using available web tools, primary sources where possible, links/citations, date awareness, synthesis, and unresolved uncertainty.

Read-only correctness review with actionable findings first, tight file/line evidence, severity, and a concise residual-risk summary.

Review trust boundaries, auth/authz, injection, secrets, filesystem/network exposure, dependencies, and exploitability without pretending a shallow lint is an audit.

Improve clarity and reduce needless complexity after behavior is understood; preserve behavior and keep cleanup separate from correctness fixes.

Create or improve codewhale skills. Use when the user wants a new skill, wants to update an existing skill, or needs guidance on when a skill should be a skill versus MCP, hooks, tools, or a plugin scaffold.

Install, update, trust, or inspect Codewhale skills from GitHub or local skill folders. Use when the user asks for available skills or wants a community skill installed.

Compatibility alias for the xlsx skill. Prefer loading xlsx.

Detect the project’s test stack, run the narrowest useful tests, create tests when authorized, and report coverage/gaps honestly.

Use when working with deepseek-v4-pro or deepseek-v4-flash in thinking mode on multi-step or plan-driven tasks. Provides rules to prevent stale references, unverified plan assumptions, and vague plan output.

Exercise the real app/API/CLI and collect observable evidence; tests alone do not count as end-to-end verification.

Start/reuse a local app, wait for readiness, inspect rendered state/console/network, act from observed selectors, and verify with evidence.

Create/edit/analyze/verify spreadsheet workbooks including XLSX/CSV/TSV.

Use before claiming CodeWhale release work is done: run the full gate sweep and list the manual QA targets.

Use to assign GitHub issues to a milestone and/or owners in bulk, verifying each.

Close resolved CodeWhale issues only after verifying the landed commit/behavior, with a positive crediting comment; never from title alone.