check-communication
Reviews messages, emails, Slack messages, and PR comments before sending — checking clarity, tone, action items, and length. Flags buried asks, accidental passive aggression, and ambiguous next steps. Make sure to use this skill whenever the user says "review this message", "check this email", "is this Slack message okay", "review my communication", "proofread this", "does this come across right", "review before sending", or pastes a draft they're about to send — even if they just ask "does this sound okay?"
What this skill does
# Communication Review
Review draft messages before sending. The goal: clear ask, right tone,
explicit next step, minimum words. Output a revised version + 2–3 specific
notes.
## The Five Checks
Run all five on every piece of communication.
### 1. Clarity — Is the point obvious in the first sentence?
If the reader has to reach paragraph 3 to understand why you're messaging
them, rewrite. The ask or context belongs at the top.
Bad: Three paragraphs of context → "Anyway, could you take a look at this?"
Good: "Can you review the auth PR by Friday? Context below."
### 2. Tone — Does it match the channel and relationship?
| Channel | Expected register |
|---|---|
| Slack DM (peer) | Casual, direct, abbreviations fine |
| Slack public channel | Slightly more formal, context for lurkers |
| Email (internal) | Professional, not stiff |
| Email (external/client) | Formal, no jargon |
| PR comment | Technical, factual, no passive aggression |
| Performance feedback | Specific, evidence-based, forward-looking |
Red flags: sarcasm that won't land in text, hedging that reads as passive
aggression, jargon with non-technical stakeholders.
### 3. Action — What do you want the reader to DO?
Vague: "Let me know your thoughts."
Clear: "Can you approve by Wednesday, or flag blockers by EOD Tuesday?"
Every message should have an explicit next step. If the action is implicit,
make it explicit.
### 4. Length — Can you cut 30%?
Usually yes. Prune:
- Context the reader already has
- Apologies for the message itself ("Sorry to bother you, but...")
- Qualifiers that add nothing ("I just wanted to quickly mention...")
- CC'd people who don't need to be CC'd
### 5. Context — Does the reader have enough to act?
Check from the reader's perspective: could they respond or take action without
a follow-up question? If not, add the missing piece.
## Auto-Detect Message Type
Before evaluating, identify the message type from content:
| Type | Signals |
|---|---|
| Status update | Progress language, timeline mentions, blockers |
| Standup | Short, daily cadence, what I did / will do |
| Proposal / RFC | Suggests a change, asks for approval, presents options |
| PR review comment | Code references, review language, suggestions |
| Disagreement | Countering a position, expressing concern |
| Slack message | Casual tone, async communication |
| Email / formal | Recipients, subject-like structure, formal tone |
State the detected type, then apply the type-specific check below in addition to the five core checks.
## Type-Specific Checks
**Status update → STATUS framework:**
- State (on track / at risk / blocked)
- Target (deliverable and deadline)
- Achieved (impact over activity — "reduced latency 40%", not "worked on caching")
- Threats (what could go wrong)
- Unblocks (what you need from others)
**Proposal / RFC → proposal structure:**
- Problem stated with real data (not vague discomfort)
- Trade-offs named — pros AND cons (listing only upsides loses trust)
- Alternatives considered and rejected with reasons
- Rollback plan present
- Open Questions section calls out decision points (silence = readers didn't know what to react to)
**PR review comment:**
- Uses prefixes: `nit:` / `suggestion:` / `question:` / `issue:` / `thought:`
- Explains WHY for blockers, not just what to change
- Avoids making the author feel dumb — explain, don't judge
**Disagreement / pushback:**
- Steel Man: does it restate the other position charitably before countering?
- "Us vs the problem" framing, not "me vs you"
- Watch for: "As I already explained...", defensive language
**Slack async:**
- Thread Rule: if it'll take >3 exchanges, call instead
- Appropriate length for the medium — Slack isn't email
## Common Issues
- **Burying the ask**: context → context → context → buried ask at line 15.
Move the ask to line 1, put context second.
- **"Per my last email..."**: almost always reads as passive-aggressive. Reword
or remove.
- **Ambiguous next steps**: "Let me know" is not an action. Specify: what, by
when, from whom.
- **Technical jargon to non-technical readers**: always flag, always simplify.
- **Apologetic opener**: "Sorry to bother you" weakens the message. Cut it.
- **No surprises risk**: flag if the message would catch stakeholders off guard
in a group setting.
## Output Format
Detected type first, then revised version, then 2–3 notes:
```
Detected type: [type]
[Revised message]
---
What's landing well: [1-2 specific strengths]
What to tighten:
- [What changed and why — specific, with a concrete rewrite if needed]
- [max 2-3 issues]
```
If the draft is mostly fine, say so briefly and note only the 1–2 small tweaks.
Peer-level tone throughout — you're reading this as a colleague who wants it to
succeed, not as a judge.
Related in Image & Video
watch
IncludedWatch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video.
physical-ai-defect-image-generation
IncludedUse when the user wants to orchestrate defect image generation, run associated setup, or handle outputs on OSMO. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment. Trigger keywords: defect image generation, dig workflow, dig pipeline, defect image detection workflow, aoi pipeline, aoi anomalygen, usd2roi anomalygen, day 0 pcba, day 1 pcba, day 1 real-photo alignment, day 1 manual roi, metal surface anomaly, glass defect, anomalygen finetune, setup_pcb, setup_metal, setup_glass, setup_pretrained, dig setup, dig datasets, dig pretrained checkpoint, dig image-edit endpoint.
accelint-react-best-practices
IncludedReact performance optimization and best practices. ALWAYS use this skill when working with any React code - writing components, hooks, JSX; refactoring; optimizing re-renders, memoization, state management; reviewing for performance; fixing hydration mismatches; debugging infinite re-renders, stale closures, input focus loss, animations restarting; preventing remounting; implementing transitions, lazy initialization, effect dependencies. Even simple React tasks benefit from these patterns. Covers React 19+ (useEffectEvent, Activity, ref props). Triggers - useEffect, useState, useMemo, useCallback, memo, inline components, nested components, components inside components, re-render, performance, hydration, SSR, Next.js, useDeferredValue, combined hooks.
elevenlabs-agents
IncludedBuild conversational AI voice agents with ElevenLabs Platform using React, JavaScript, React Native, or Swift SDKs. Configure agents, tools (client/server/MCP), RAG knowledge bases, multi-voice, and Scribe real-time STT. Use when: building voice chat interfaces, implementing AI phone agents with Twilio, configuring agent workflows or tools, adding RAG knowledge bases, testing with CLI "agents as code", or troubleshooting deprecated @11labs packages, Android audio cutoff, CSP violations, dynamic variables, or WebRTC config. Keywords: ElevenLabs Agents, ElevenLabs voice agents, AI voice agents, conversational AI, @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, @elevenlabs/agents-cli, elevenlabs SDK, voice AI, TTS, text-to-speech, ASR, speech recognition, turn-taking model, WebRTC voice, WebSocket voice, ElevenLabs conversation, agent system prompt, agent tools, agent knowledge base, RAG voice agents, multi-voice agents, pronunciation dictionary, voice speed control, elevenlabs scribe, @11labs deprecated, Android audio cutoff, CSP violation elevenlabs, dynamic variables elevenlabs, case-sensitive tool names, webhook authentication
humanizer
IncludedHumanize AI-generated text by detecting and removing patterns typical of LLM output. Rewrites text to sound natural, specific, and human. Uses 28 pattern detectors, 560+ AI vocabulary terms across 3 tiers, and statistical analysis (burstiness, type-token ratio, readability) for comprehensive detection. Use when asked to humanize text, de-AI writing, make content sound more natural/human, review writing for AI patterns, score text for AI detection, or improve AI-generated drafts. Covers content, language, style, communication, and filler categories.
generating-mermaid-diagrams
IncludedSalesforce architecture diagrams using Mermaid with ASCII fallback. Use this skill when generating text-based diagrams for Salesforce architecture, OAuth flows, ERDs, integration sequences, or Agentforce structure. TRIGGER when: user says "diagram", "visualize", "ERD", or asks for sequence diagrams, flowcharts, class diagrams, or architecture visualizations in Mermaid. DO NOT TRIGGER when: user wants PNG/SVG image output (use generating-visual-diagrams), or asks about non-Salesforce systems.