メニュー
Use when triaging an inbox: fetching unprocessed emails, classifying each into a label taxonomy, and applying labels via the provider's API. Provider-agnostic skeleton; scripts/label.py must be wired to your email provider before first use.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when a question has a factual or best-practice answer but a single response might hallucinate, or when the user asks to "get consensus" or "sanity check this with multiple agents". Spawns N subagents on the same question with varied personas and tallies answer frequency.
Use when a question has real trade-offs and no single right answer: architecture choices, library selection, design direction. Runs a multi-round debate where agents see each other's positions across rounds, converge, and a synthesis is produced. For factual questions, use consensus instead.
Use when a workflow built on an MCP server has stabilized and should become a production skill, or when the user says "convert this MCP workflow to a skill". Produces a skill with a bundled script that calls the underlying API directly, eliminating the per-session MCP token overhead.
Use when a PRD or spec exists and needs to become an executable backlog. Breaks the document into GitHub issues structured as vertical-slice tracer bullets, marks each as HITL or AFK, and files a final QA issue. Requires the gh CLI.
Use when implementing any task that should end in a commit. Encodes the definition of done: understand, implement with red-green-refactor for backend logic, run typecheck and tests until clean, then commit. Also the unit of work for autonomous loops.
Use when creating or revising any Claude Code skill, or when the user says "write me a skill for X". Encodes the house conventions: frontmatter style, body structure, when to bundle scripts, and project vs user scope.
| name | email-labeler |
| description | Use when triaging an inbox: fetching unprocessed emails, classifying each into a label taxonomy, and applying labels via the provider's API. Provider-agnostic skeleton; scripts/label.py must be wired to your email provider before first use. |
Inbox triage as a skill. The script handles fetching and labeling (deterministic); you handle classification (judgment). This is the Lab 3 artifact from Claude Code 101, stripped to a provider-agnostic skeleton.
Setup required: scripts/label.py has TODO markers for provider auth and API calls. Wire it to Gmail, Outlook, or Fastmail once; the skill works forever after. Ask Claude to do the wiring: "implement the TODOs in scripts/label.py for Gmail using the Gmail API".
Four labels. Edit to taste; the taxonomy is the part that's yours.
| Label | Meaning |
|---|---|
action-required | A human must do something: reply, decide, pay, sign |
waiting-on | You're blocked on someone else; their reply closes it |
reference | Keep: receipts, confirmations, documents worth finding later |
newsletter-ish | Bulk content. Read never or in batches |
python scripts/label.py fetch --max 50. Returns JSON: id, sender, subject, snippet per email.action-required; an automated confirmation = reference; anything with an unsubscribe link = newsletter-ish unless it asks for action. When genuinely ambiguous, prefer action-required; a false alarm beats a missed deadline.python scripts/label.py apply --input labels.json where labels.json maps email id to label.action-required subjects so the user sees what needs them.Never delete or archive anything. Labeling only. Never label more than the fetched batch in one run.
Once trusted, run it on a schedule: a /loop in a session, or a scheduled task invoking this skill daily. Start supervised for the first few runs; spot-check 10 classifications before going hands-off.