本文へ移動
cccskills
無料GitHub で公開

Inbox Triage — sort your inbox into reply-now / later / ignore

Triage an email inbox — "triage my inbox", "sort my inbox", "what emails need a reply", "clear my inbox". Governs the Gmail or Workspace connector you already have rather than shipping its own integration.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md4.7 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Inbox Triage

Does: sorts your inbox into reply-now / reply-later / ignore, and drafts the urgent replies for your approval. Solves: an inbox so full you don't know where to start. Triggers on: "triage my inbox" · "what emails need a reply"

How Maxim does this (consume the connector — never rebuild it)

Maxim does not ship its own Gmail integration. It uses the email connector you already have — the Anthropic Gmail / Google Workspace connector (vetted, GA, all users) or any email MCP you've added — and adds the layer the raw connector lacks: a triage framework, a privacy gate, and confidence. (ADR-018 + ADR-012: Maxim is the governed brain; the connector is the hands.)

Per-surface availability (verified 2026-06-26):

  • Claude Desktop / web: turn on the Google Workspace → Gmail connector (+ → Manage connectors). Works immediately.
  • Claude Code: the Anthropic Gmail connector authenticates at claude.ai → Settings → Connectors, then appears in Claude Code automatically; or claude mcp add a Gmail MCP (use --scope user so it's available everywhere, including scheduled runs).

What it does, step by step

  1. Pull recent unread via the connector (e.g. is:unread newer_than:1d). Read-only.
  2. Classify each message — reply-now / reply-later / ignore — using the Eisenhower matrix (urgency × importance), with the sender + ask + deadline as signals.
  3. Draft replies for the reply-now set. The connector is draft-only and cannot send — drafts wait in your Gmail for one-tap manual send. This is a platform-enforced safety, on top of Maxim's own.
  4. Return a digest: counts per bucket + the drafts awaiting you + the 1–2 that actually matter today.

Behavioral + governance overlay

  • Frameworks (cited per ADR-007): Eisenhower Matrix (urgency × importance) · Fogg Behavior Model (surface the easiest high-value reply first to beat inbox paralysis).
  • CSO auto-loop (non-negotiable): email is personal data. security-analyst gates the run — message bodies/addresses are never copied into logs or external output; the RunLog stores counts and message-ids, not content.
  • No auto-send, ever: drafts only (the connector enforces it; Maxim's dry-run reinforces it).
  • No-fabrication: it triages only what the connector returns; it never invents senders, deadlines, or "urgent" flags.
  • Confidence tag (ADR-010) on the triage — 🟡 if the connector isn't connected (it'll tell you how to connect, not guess).

The unattended "every morning" version

The same skill, wrapped by mxm-orchestrator (ADR-022) as a scheduled workflow: runs each morning via the usage-aware-scheduler, dry-run by default, bounded by a budget guard, idempotent (won't re-draft the same thread), and draft-only end to end. Definition: orchestrator/workflows/inbox-triage.mjs. Go-live requires a connected email tool + explicit operator approval (/mxm-workflow go-live inbox-triage). For headless/scheduled runs, use a user-scoped Gmail MCP (interactive Desktop connectors assume the app is open).

Output

A triage digest — reply-now (with drafts ready) · reply-later · ignore — counts + the 1–2 that matter most. Nothing sent.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

AI video and image prompt engineering — video generation, image generation, product video, social media video, cinematic direction, animation, Seedance, Higgsfield, GPT-Image-2. Applies 15 Maxim-authored cinematic styles and the atomic prompt schema.

日本語の概要は準備中です。原文の説明を表示しています。

DrNabeelKhan/maxim22026年9月17日 更新

Apply a named behavioral science framework to any task — persuasion, nudge design, habit formation, conversion, influence, negotiation, cognitive bias, choice architecture, decision framing. The framework backbone every other Maxim skill loops into for psychological validation.

日本語の概要は準備中です。原文の説明を表示しています。

DrNabeelKhan/maxim22026年9月17日 更新

Founder and CEO automation — morning standup, overnight cycle, burn rate, runway, pipeline review, investor and fundraising prep, executive summary, daily digest. Operates on the .mxm-executive-summary context.

日本語の概要は準備中です。原文の説明を表示しています。

DrNabeelKhan/maxim22026年9月17日 更新

Tear down a company on request — "do a teardown of", "break down this company", "analyze this competitor's business". Fixed-template output: business model, how they make money, positioning, and where they are exposed.

日本語の概要は準備中です。原文の説明を表示しています。

DrNabeelKhan/maxim22026年9月17日 更新

Track what a competitor actually shipped — "watch this competitor", "track what they ship", "what changed at". Compares against the last snapshot and reports only real moves: launches, pricing changes, positioning shifts.

日本語の概要は準備中です。原文の説明を表示しています。

DrNabeelKhan/maxim22026年9月17日 更新

Regulatory review across 14 frameworks — compliance review, privacy impact assessment, regulatory gap analysis, data subject rights, consent mechanisms, audit trail, AI ethics review, data flow mapping. Covers GDPR, PIPEDA, HIPAA, PCI-DSS, SOC2, UAE-PDPL, EU AI Act and more.

日本語の概要は準備中です。原文の説明を表示しています。

DrNabeelKhan/maxim22026年9月17日 更新

DrNabeelKhan のスキルをすべて見る

このスキルの問題を報告する