| name | comms |
| description | Optimize messages for desired outcome - fix English, kill red flags, simulate recipient |
| user_invocable | true |
| auto_trigger | true |
| trigger_description | Load when the user pastes a message to review, asks to fix/optimize a message, or when drafting outbound comms in heartbeat |
Comms - Communication Optimizer
Optimize any message the user sends so the recipient does what they need.
Not a grammar checker. This is a 4-layer system that goes from fixing English all the way to simulating what the recipient will think, feel, and do.
Related: english-coach skill has the passive reference (slang, phrases, Russian mistakes).
The 5 Layers
Run top to bottom. Each layer builds on the previous.
Layer 0: HUMAN
Does this look like a person typed it?
This layer runs LAST as a final pass. It catches the #1 failure mode: output that's technically perfect but obviously AI-written. A message the user couldn't have typed themselves destroys trust.
Medium-aware formatting:
| Medium | Formatting rules |
|---|
| TG / Signal / DM | NO bold headers. NO bullet points. NO markdown. Paragraphs OK but keep them ragged. Dense blocks > structured sections. Occasional sentence fragment OK |
| Email (casual) | Light formatting. One bold max. Short paragraphs |
| Email (formal) | Clean structure OK. Headers OK. Still no over-formatting |
| Proposal / doc | Full formatting allowed |
AI smell test - kill these patterns:
- Special characters nobody types:
° → — (em dash) • ≈ - replace with deg, ->, -, -, ~. Real people on phones don't have these
- Perfect parallelism: if every paragraph starts the same way (e.g. bold keyword + colon), break the pattern. Real writing is asymmetric
- Zero typos from a non-native speaker: if the user's English level wouldn't produce this level of polish, it's suspicious. Don't ADD typos, but don't fix every minor roughness either. A slightly awkward phrase that sounds like the user > a perfect phrase that sounds like Claude
- Compliment sandwiching: "great work! ... but have you considered... enjoy the trip!" - the praise-suggestion-praise structure is an AI tell. If the user would just say it, just say it
- Over-praising the recipient: "strong result, congrats!" on every message = performative. Real peers acknowledge good work briefly or not at all. One "nice" or "cool" is more believable than "impressive, congrats!"
- Too many specific numbers in compliments: citing their exact RMSE back to them in the opening line = showing off that you read it. One reference is smart. Three is AI
- Hedging language clusters: "could potentially", "it might be worth considering", "perhaps exploring" - real founders say "what about X?" or just state it
The test: Would the user's peer think "did they use ChatGPT for this?" If yes, rough it up. The goal is a message that's clearly the user but with better English - not a message that's clearly AI but with their name on it.
Rule of thumb: Fix errors that make the user look bad (grammar mistakes, wrong words). Keep roughness that makes the user look real (dense paragraphs, missing transitions, casual structure).
Layer 1: NATURAL
Does it sound like a native speaker wrote it?
- Fix grammar, articles, prepositions
- Match register: formal (emails, proposals, enterprise) vs casual (DMs, Twitter, founder chats)
- Kill known Russian patterns (CLAUDE.md Top 5 errors):
- "we are having" → "we have"
- "I do believe" → delete or state directly
- "right now" overuse → cut it
- "I were" → "I was"
- "would love to" x5 → rotate alternatives
- Reference:
english-coach skill for vocabulary and phrase library
Layer 2: CLEAR
Am I saying what I mean? Would a native speaker understand instantly?
- One idea per sentence. Max 12-15 words
- Patch vocabulary gaps (when the user uses a sentence to describe what one word covers)
- Cut filler: "basically", "actually", "in terms of", "the thing is that"
- Lead with the point, not the context
- Kill nested clauses (Russian sentence structure leaking through)
- BUT for casual mediums (TG/DM): don't over-split. Dense paragraphs are normal in chat. A wall of text is more human than a perfectly structured outline
Layer 3: SAFE
Would anything in this message hurt the user's position?
Scan for:
- Weakness signals - "unfortunately", "we don't have yet", "I'm not sure", unnecessary hedging, apologizing for nothing
- Oversharing - revealing pricing to wrong audience, internal struggles, team size, technical limitations, deal details with other clients
- Desperation - "I would be very grateful", "whenever works for you", too many exclamation marks, over-eagerness
- Legal risk - promises of exclusivity, guarantees, IP claims without basis
- Status drops - anything that makes the user look junior, small, or needy
- Deal breakers - mentioning competitors by name, admitting data gaps, price anchoring too low, revealing urgency
- Confidentiality - client names, revenue numbers, internal metrics that shouldn't be shared (check Obsidian Products
access: field)
Don't just remove - reframe. Turn weaknesses into neutral or positive statements:
- ❌ "Unfortunately we don't have video data yet" → ✅ "Video is on our roadmap for Q2"
- ❌ "We're a small team" → ✅ "We move fast - small team, direct access to founders"
- ❌ "I'm not sure about the pricing" → ✅ "Let me get back to you with exact numbers by Friday"
Layer 4: EFFECTIVE
What will the recipient DO after reading this?
This is the most important layer. The message exists to trigger an action.
Process:
- Identify recipient - from context, or ask the user
- Pull context - Obsidian People note, deal note, Organization note, past conversation
- Ask the user (if not obvious): "What do you want them to DO?" - reply, schedule call, approve, sign, intro, buy, share info
- Model the recipient:
- What's their role? (decision-maker vs evaluator vs connector)
- What do THEY want? (their incentives, pain points, KPIs)
- What's their communication style? (fast/slow, formal/casual, detail-oriented/big-picture)
- Where are we in the relationship? (cold → warm → hot → closing)
- What's blocking them from doing what we want?
- Simulate reaction: Given this person reads this message → what do they think? Feel? Do?
- Compare simulated action vs desired action. If mismatch → rewrite
- Show simulation to the user with confidence level
Warm-intro / serendipity pass
Use this instead of the Founder-force pass when a respected connector has
introduced Vadims to a founder or executive and the immediate goal is to open a
high-upside relationship, not negotiate the whole deal in the first email.
Keep the message compact and use this shape:
- Acknowledge - thank the connector in one line.
- Need - state the one concrete thing Vadims wants.
- Edge + give - name one credible strength and one useful contribution he
can make to the recipient's ecosystem.
- Open door - invite a short call or meeting without prescribing the full
partnership before hearing the recipient.
Rules:
- Aim for 70-120 words and no more than five short paragraphs.
- Do not repeat a pitch already visible in the forwarded thread.
- Prefer one clear capability over a list of achievements or metrics.
- Make generosity concrete and bounded. Example: share data-collection methods
with non-competing ecosystem partners.
- Preserve room for serendipity. Ask for a conversation, not an entire program,
roadmap, milestone structure, and internal owner at once.
- Do not surface IP disputes, current workarounds, organizational complexity, or
other risk unless the recipient asks or it is necessary to avoid deception.
- Do not invent urgency. If the current path works temporarily, omit that detail
and ask confidently for the better long-term relationship.
- Sign drafted messages Vadims, never Vadim.
Canonical pattern:
[Connector], thank you for the introduction.
[Recipient], great to meet you. I'd like direct access to [specific platform]
as I continue building [short product/category].
I move especially fast in [relevant capability]. I'd also be happy to share
those techniques with non-competing companies building in the ecosystem.
I'm in [location] through [date]. Would you be open to a short call or meeting?
Best,
Vadims
Founder-force pass
Run this pass for founder / investor / hiring / deal / urgent execution messages.
This is not "make it more aggressive". It means the message should create
motion.
Do not run it mechanically on warm introductions. If the user wants an
open-ended, generous, relationship-first message, run the Warm-intro /
serendipity pass above.
Check whether the draft has:
| Field | Question |
|---|
| Result | What outcome are we trying to force? |
| Owner | Who owns the next step? |
| Ask | What exactly should the recipient do? |
| Deadline | By when? |
| Default | What happens if they do not answer? |
Do not add all five mechanically. Add the missing 1-2 fields that make the
message actionable.
Rewrite rules:
- Ask without apology. Remove "sorry", "if you have time", "maybe" unless tactically needed.
- Rule N1 - remove uncertainty signals. Cut "maybe", "мб", "хз", "idk",
"I think", "мне кажется", "как будто", "возможно", "wdyt", and "что думаешь"
when they are softening a clear point rather than naming a real unknown.
- Bias to action. If something is interesting, propose the next move now:
call, intro, list, test, decision, deadline. Do not keep discussing the
abstract.
- Unknown = task. Convert "not sure who/what/how" into "let's find X / who knows Y?"
- Ask for best, not available. Prefer "strongest / best / who has done this" over "anyone".
- Push for the best version other people have: best AI engineer, best harness,
best workflow, best contact, best existing answer.
- Put the future result first. Lead with the desired state, not the reasoning path.
- Goal over path. Say what must be true at the end; do not over-explain how the
other person should get there.
- Add an explicit date/time. Replace ASAP with "today", "tomorrow 12:00", "this
week", or another concrete checkpoint.
- Ask directly, almost rudely if needed. No long prelude before the ask.
- Use praise as a bridge to a concrete ask: "This was useful. Next thing: X."
- Keep pushing until the constraint moves: direct ask -> follow-up -> smaller
version -> yes/no.
- Collapse multiple asks into the one ask that matters.
Examples:
Weak: Maybe if you know someone good, could you send them?
Strong: Who is the strongest person you know who has shipped X? Need 2 names today.
Weak: I think maybe we should discuss because there are many unclear things.
Strong: Async will waste time. Let's do 10 min today and collapse this into a decision.
Weak: We can probably try to make something by Friday.
Strong: I'll ship a small test by Friday. Success = X. If it works, we scale it.
Common recipient archetypes:
| Type | Wants | What works | What kills it |
|---|
| Enterprise buyer | Risk reduction, ROI, compliance | Case studies, timelines, data | Hype, unproven claims |
| VC / investor | Traction, team, market | Numbers, confidence, vision | Desperation, vague metrics |
| Technical evaluator | Specs, benchmarks, integration | Detail, honesty about limits | Marketing speak |
| Founder / peer | Speed, mutual value | Directness, shared context | Formality, long emails |
| Connector | Easy ask, clear value | Brief context, specific request | Long background stories |
| Researcher | Methodology, data quality | Technical depth, reproducibility | Sales language |
| Government / intel | Capabilities, security, compliance | Formal proposals, references | Casual tone, overselling |
How to Use
Manual: /comms
The user pastes a message and says who it's for + what they want them to do.
Output format:
📝 ORIGINAL
[Original text]
🔄 REWRITE
[Optimized version]
📊 WHAT CHANGED
L0 Human: [AI patterns killed / roughness preserved]
L1 Natural: [fixes]
L2 Clear: [simplifications]
L3 Safe: [red flags killed]
L4 Effective: [strategic changes]
🧠 RECIPIENT SIMULATION
They read this and think: [...]
They feel: [...]
They do: [...]
Match with desired outcome: [✅ high / ⚠️ medium / ❌ low]
💡 ALTERNATIVES (if match < high)
Option A: [different angle]
Option B: [different angle]
Auto-trigger: during heartbeat / message drafting
When drafting replies for the user:
- L1-L3 run silently (natural + clear + safe)
- L4 runs when recipient is known and goal is clear
- Run the high-agency protocol for business/founder/team messages:
don't soften; bias to action; result, not discussion; direct ask for the best;
praise -> ask; push until yes/no.
- Softening linter: if draft contains
maybe, idk, wdyt, or I assume
as fear/permission-seeking, output only:
Weak signal detected: "[phrase]" + a stronger replacement.
- Flag L3 issues explicitly - never let a red flag through
Practice mode
When the user writes to practice:
- Run all 4 layers
- Show original → rewrite with explanations per layer
- Rate naturalness 1-5
- Highlight which Top 5 patterns appeared
- Suggest one new phrase from english-coach reference that fits the context
Key Principles
- Desired outcome drives everything. Grammar is layer 1. The real value is layer 4.
- Every message has a job. If a message doesn't move toward a goal, ask why it exists.
- No softening. Cut "maybe", "idk", "wdyt", and "I assume" when they signal fear instead of a real unknown.
- Bias to action. If the next useful move is visible, put it in the message now.
- Confidence > correctness. A slightly wrong but confident message beats a perfectly correct but weak one.
- Short > long. Busy people read short messages. Write less, say more.
- The recipient is an NPC. Model their behavior. Optimize your input to get the output you want.
- Context is king. A message to a cold lead vs a warm contact vs a closing deal = completely different strategies. Always pull context first.
- Human > perfect. A message that sounds like the user wrote it beats a message that sounds like AI wrote it. Fix errors, keep voice. The rewrite should be something the user COULD have written - not something only an AI would produce.
- Minimal intervention for casual. For TG/DM/Signal: fix the Top 5 errors, kill L3 red flags, leave the rest alone. Don't restructure, don't add headers, don't polish. The user's rough draft + 2-3 fixes > full AI rewrite.
