메뉴
Lightweight inbox triage across all configured email accounts. Dispatches fast subagent to fetch, then reviews with judgment. Use when user says "check my inbox", "any new emails?", "check email". See also: `draft-reply` once you pick an email to respond to.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Iteratively improve a local artifact (draft, document, page) by running a hill-climbing loop. The user names the artifact, the evaluator, and the budget. Claudia edits the artifact, scores it, keeps it if better or reverts if worse, repeats. Use when user says "iterate on this", "loop on this until it's better", "run experiments on this draft", "auto-research this", "make this better and don't stop until it's good". Workspace-scoped, never touches live files, no external actions during the loop.
Propose a personalized team of specialized agents based on the user's profile, goals, and how they actually work. Runs the proposal through an independent Checker, gates on the user's approval, and applies with rollback. Use when the user says "build my team", "set up my agents", "what agents should I have", "design my team", "/build-team", or asks Claudia to tailor her team to their work. See also: `hire-agent` for proactive single-agent suggestions; `structure-generator` for folder scaffolding; `capability-suggester` for command-level additions.
End-of-session reflection generating persistent learnings about user preferences, communication patterns, and cross-session insights. Activates when wrapping up a session, or when user says "let's wrap up", "end the session", "time to reflect", or "meditate". Also extracts judgment rules from decisions made during the session. See also: `weekly-review` and `growth-check` for time-based reflection cadences.
Launch the Brain Monitor TUI, a real-time terminal dashboard for watching Claudia's memory system. Triggers on "brain monitor", "show dashboard", "memory dashboard", "terminal brain". See also: `brain` for a 3D graph view in the browser.
Launch the Brain Visualizer, a real-time 3D view of memory and relationships. Triggers on "show your brain", "visualize memory", "open the brain", "memory graph". See also: `brain-monitor` for a terminal dashboard alternative.
Process meeting notes or transcript to extract decisions, commitments, and insights. Use when user shares transcript or says "capture this meeting", "here are my notes from the call". See also: `meeting-prep` for pre-call briefings; `follow-up-draft` for post-meeting emails.
| name | inbox-check |
| description | Lightweight inbox triage across all configured email accounts. Dispatches fast subagent to fetch, then reviews with judgment. Use when user says "check my inbox", "any new emails?", "check email". See also: `draft-reply` once you pick an email to respond to. |
Two-tier inbox triage: fetch fast, then judge what matters.
/inbox-checkThis skill uses a two-tier approach to save tokens and time:
Check which email MCP tools are available:
gmail.* or similar)outlook.* or similar)If no email tools available:
"I don't have access to any email accounts. Would you like to set up an email integration? Check /connector-discovery for options."
Use the Task tool to dispatch a Haiku subagent:
Subagent prompt:
"Check all available email accounts for recent messages (last 24 hours or since last check).
For each email, extract:
- From (name and email)
- Subject
- Date/time
- Brief summary (1 sentence)
- Has attachments? (yes/no)
- Seems urgent? (yes/no, based on subject and sender)
Return as a structured list, grouped by account if multiple.
Do not read full email bodies unless the subject is unclear.
Limit to 20 most recent per account."
Once the fetch results come back, apply judgment using your knowledge of the user's relationships, priorities, and patterns:
Categorize each email:
| Category | Criteria |
|---|---|
| Needs Reply | From known contacts, asks a question, requires action |
| Worth Reading | Relevant to active projects, from important people |
| Can Wait | Newsletters, notifications, low-priority updates |
| Skip | Marketing, spam that got through, irrelevant notifications |
Use relationship context:
people/, note the relationship context**📬 Inbox Check**
### Needs Reply
- **[Name]** — [Subject] ([time ago])
[1-line context: what they need, any relationship notes]
### Worth Reading
- **[Name]** — [Subject] ([time ago])
[Why it matters]
### Can Wait
- [count] emails (newsletters, notifications, low-priority)
[List briefly if < 5, otherwise summarize]
---
Want me to:
- Draft a reply to any of these?
- Read the full text of something?
- File any of these to memory?
Quick, scannable, opinionated. The whole point is to save the user from inbox overwhelm. Be decisive about what matters and what doesn't.