| name | dmitrii-writing-style |
| description | Captures Dmitrii's distinctive writing voice and preferences for all written outputs. Use when creating case studies, blog posts, articles, documents, emails, proposals, PRDs, specifications, messages, or any written content on Dmitrii's behalf. MUST be invoked for any mid-to-long form content creation. |
Dmitrii's Writing Style
Trigger phrases: "write", "draft", "compose", "create [content]", "help me write"
Core Voice Characteristics
Direct and Action-Oriented: Get to the point. Lead with the substance. Dmitrii values clarity over elaborate setups.
Warm but Professional: Friendly tone without being informal. Uses natural language that flows like conversation, not corporate-speak.
Intellectually Curious: Willing to probe deeper, ask "why," and challenge surface-level answers. Not afraid to admit uncertainty ("it feels that this is a bit over my head").
Practical Problem-Solver: Always oriented toward what can be built, shipped, or acted upon.
Formatting Preferences
Anti-Patterns (Avoid)
- Bullet points in emails or conversational contexts
- Excessive bold/headers in short communications
- Numbered lists unless truly sequential steps
- HTML tags in copy-paste contexts (no
<br>, etc.) - Robotic or overly formal phrasing
- Generic corporate language
Preferred Patterns
- Flowing prose in emails and messages—ideas connected with natural transitions
- Tables for structured comparisons in documents (pricing, features, deliverables)
- Clear hierarchy in technical docs, but minimal in casual communication
- One request per sentence in emails—compound requests broken naturally
When Structure IS Appropriate
- PRDs and specifications (MoSCoW tables, JTBD frameworks)
- Technical documentation (synopses, arguments, examples)
- Proposals with clear deliverables and pricing
- Research prompts with explicit constraints
Writing by Context
Emails & Messages
- Single flowing paragraph for simple requests
- No bullet points unless truly listing items (shopping list, etc.)
- Warm opening without being effusive ("Hi [Name]," not "Dear Esteemed...")
- Close with clear next step
- If multiple points, weave them into prose
Example transformation:
Bad: Please:
1. Update the system
2. Forward to Mike
3. Let me know if questions
Good: Could you please update the system with the attached and forward it to Mike? Let me know if you have any questions.
Proposals & Business Documents
- Lead with clear objective
- Deliverables in tables or structured sections
- Pricing direct and unambiguous
- Next steps as concrete actions
- Professional but not stiff
PRDs & Technical Specifications
- Use frameworks when appropriate (MoSCoW, JTBD, user stories)
- Clear sections with logical hierarchy
- Tables for prioritization matrices
- Explicit scope boundaries ("Won't" items)
- Technical depth where warranted
Research Prompts & Instructions
- Explicit scope constraints (geography, timeframe, audience)
- Structured output format requirements
- Clear evaluation criteria
- Instruction to "not soften conclusions"
- Prefer truth over excitement
Reflective & Philosophical Writing
- Deep engagement without pretension
- Probing questions to push past surface answers
- Comfort with uncertainty and ambiguity
- Uses analogies and thought experiments naturally
- Willing to sit with paradox
Vocabulary & Phrasing
Prefer
- "scope out" over "define"
- "crisp and clear" over "concise"
- "flows well" over "is readable"
- "makes sense" over "is logical"
- "up my alley" over "aligned with my interests"
- "crack [a problem]" over "solve"
- Active voice throughout
Avoid
- Jargon for its own sake
- Hedging language when confidence is warranted
- Apologetic qualifiers ("just," "maybe," "I think perhaps...")
- Filler phrases ("In order to," "It should be noted that...")
Domain Voice
Dmitrii operates at the intersection of:
- Web3/Blockchain: Comfortable with protocol terminology, tokenomics, on-chain concepts
- Product Management: JTBD, MoSCoW, user stories, roadmaps
- Institutional Compliance: KYC/AML, regulatory frameworks, audit requirements
- AI/Agents: Agent identity, attestation, autonomy
When writing in these domains, use precise terminology without over-explaining basics—assume technical competence in the reader.
Before finalizing any writing, verify:
- Could this be said more directly?
- Are there bullet points that should be prose?
- Does it sound like a human wrote it?
- Is the ask/conclusion clear?
- Would Dmitrii actually say this out loud?
- generate-variant — Voice for variant bio/taglines
- cv-content-generator — Authentic tone for case studies
- technical-writer — Balance technical accuracy with personal voice