.claude/skills/collaborative-writing/SKILL.md
Co-create any professional writing through Socratic dialogue using established voice patterns
npx skillsauth add ajbcoding/claude-skill-eval collaborative-writingInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Guide professional and institutional writing projects from conceptual intent to polished draft through Socratic questioning and collaborative development. This skill integrates your established voice patterns and language preferences (when available) to ensure authentic, effective communication.
Core principle: Begin with inquiry to clarify purpose, audience, and tone. Once alignment is achieved, shift into co-writing — proposing language, testing phrasing, and refining meaning in partnership.
Lexicon Dependencies (enhance quality when available, not required):
~/lexicons_llm/03_narrative_patterns.md (for structure and storytelling patterns)~/lexicons_llm/04_language_bank.md (for voice consistency and preferred language)~/career-applications/[job-slug]/01-job-analysis.md (if writing is job-related)User-Provided:
"I'm using the Socratic-Collaborative Writing Skill to refine and co-develop your message."
First, gather any existing materials:
Ask for rough writing:
Do you have any rough writing you'd like to use?
This could be:
- Draft paragraphs or fragments
- Earlier versions you want to refine
- Stream-of-consciousness notes
- Any text you've already started
If yes, please share it. If no, that's perfectly fine — we'll develop from scratch.
Ask for bullet points or ideas:
Do you have any bullet points or ideas you'd like to include in the narrative?
This could be:
- Key points you want to make
- Important details or examples
- Specific phrases or concepts
- Ideas you want to weave in
Please share any bullets or ideas you have. If none yet, we'll discover them through dialogue.
Capture user inputs:
Determine writing context:
Ask user: "What type of writing are we developing together?"
If job-related writing:
Ask: "What is the job slug (directory name) for this application?"
~/career-applications/[job-slug]/Attempt to load lexicons:
Try: ~/lexicons_llm/03_narrative_patterns.md
Try: ~/lexicons_llm/04_language_bank.md
Try: ~/career-applications/[job-slug]/01-job-analysis.md
Report status clearly:
✓ Lexicons loaded:
- Narrative patterns: [brief description of what's available]
- Language bank: [brief description of what's available]
- Job analysis: [if available]
I'll reference your established voice patterns throughout the process.
OR
ℹ Lexicons not found (this is fine!)
We'll develop your writing through Socratic dialogue alone.
I'll help you discover and articulate your authentic voice through our conversation.
If general professional writing:
Attempt to load lexicons:
Try: ~/lexicons_llm/03_narrative_patterns.md
Try: ~/lexicons_llm/04_language_bank.md
Report status clearly:
✓ Lexicons loaded:
- Narrative patterns: Found your established storytelling patterns
- Language bank: Found your preferred language and phrases
I'll reference these to maintain voice consistency.
OR
ℹ No existing lexicons found (this is perfectly fine!)
We'll collaborate to develop your message from scratch.
The Socratic process will help us discover the right voice together.
Key principle: Lexicons are helpful but NOT required. Proceed confidently with or without them.
Goal: Establish shared clarity on purpose, audience, and context.
Method: Ask one question at a time — often multiple choice — to uncover:
Examples:
Examples:
Examples:
Examples:
Output: Intent Statement
After discovery dialogue, synthesize:
## Intent Statement
Purpose: [Primary goal of the communication]
Audience: [Primary and secondary audiences]
Desired Response: [What you want readers to think/feel/do]
Tone: [Emotional register and voice stance]
Constraints: [Key considerations: length, approvals, sensitivities]
Context: [Timing and situational factors]
Does this capture your intent accurately?
Wait for user confirmation before proceeding.
Transition cue: "Shall we move from discovery into message shaping?"
Goal: Identify and organize the core ideas and their relationships.
Method: Continue the Socratic process to establish message hierarchy.
Examples:
Examples:
Examples:
Output: Message Map
After architecture dialogue, synthesize:
## Message Map
Core Ideas:
1. [First essential idea]
- Supporting evidence: [examples/proof]
- Why it matters: [significance]
2. [Second essential idea]
- Supporting evidence: [examples/proof]
- Why it matters: [significance]
3. [Third essential idea, if applicable]
- Supporting evidence: [examples/proof]
- Why it matters: [significance]
Narrative Arc: [How ideas connect and flow]
Emotional Journey: [How reader's perspective should shift]
Call to Action: [Specific request or desired response, if applicable]
Does this structure capture your message effectively?
Wait for user confirmation before proceeding.
Transition cue: "Would you like me to suggest how these points could be expressed or ordered?"
Goal: Define the emotional texture, voice, and stylistic stance.
Method: Explore through contrast and testing.
Ask questions that reveal preferences:
Test voice dimensions:
Examples:
Explore phrasing preferences:
If narrative_patterns.md and language_bank.md are available:
After initial voice exploration, provide enhanced analysis:
## Voice Calibration: Enhanced with Your Lexicon
Based on our conversation and your established patterns:
### From Your Narrative Patterns (narrative_patterns.md):
- **Opening strategies you favor:** [List patterns found]
- **Evidence presentation:** [How you typically structure proof]
- **Narrative rhythm:** [Sentence structure observations]
- **Recurring structures:** [Frameworks you use consistently]
### From Your Language Bank (language_bank.md):
- **Preferred action verbs for this context:** [List 8-12 verbs]
- **Power phrases you've used successfully:** [List 4-6 phrases]
- **Industry/field-specific terms:** [Relevant terminology]
- **Phrasing patterns:** [How you typically construct key ideas]
### Recommended Voice for This Piece:
[Synthesis of:]
1. Audience needs from Intent Statement
2. Message requirements from Message Map
3. Your authentic patterns from lexicons
4. Voice preferences from our exploration
This approach will sound like you while serving your strategic intent.
Does this voice profile feel authentic and appropriate?
If lexicons NOT loaded:
Provide voice summary based purely on dialogue:
## Voice Profile
Based on our conversation, here's the voice I'm hearing:
Tone: [Description based on user's choices]
Rhythm: [Sentence structure preferences]
Stance: [Relationship to reader]
Language level: [Formality, technical depth]
Emotional register: [Where on key dimensions]
Key voice principles:
- [Principle 1 from user's preferences]
- [Principle 2 from user's preferences]
- [Principle 3 from user's preferences]
Does this feel right for your message?
Output: Voice Profile
Wait for user confirmation.
Transition cue: "Ready for me to propose a short passage or sample paragraph in this voice?"
Goal: Move from reflection to writing.
Method: Co-create the text based on agreed message framework.
Explain the process:
I'll draft this in small segments (50-150 words at a time) so we can refine together.
After each segment, I'll ask:
- "Does this feel aligned with your voice?"
- "Would you like me to adjust tone, focus, or rhythm?"
- "Should we try a different approach to this section?"
We'll build the complete draft iteratively, keeping your authentic voice throughout.
For each section of the Message Map:
Draft a segment (50-150 words)
Present the draft segment:
## [Section Name] - Draft Segment
[50-150 words of drafted text]
Does this feel aligned with your voice and intent?
Would you like me to:
- Adjust the tone (more/less formal, warm, direct)?
- Change the focus or emphasis?
- Try a different rhythm or phrasing?
- Keep it as-is and continue?
If lexicons loaded: Voice Consistency Check
After each segment, verify against language bank:
Quick consistency check against your lexicon:
✓ Action verbs used: [list verbs from this segment]
- From your language bank: [which ones match user's patterns]
- New verbs introduced: [note any that aren't in lexicon]
✓ Phrasing patterns: [note alignment with narrative_patterns.md]
✓ Rhythm check: [does sentence structure match user's natural flow]
Does this maintain your authentic voice?
Incorporate feedback iteratively
Continue to next segment
After completing 2-3 segments:
## Mid-Draft Review
We've developed:
- [Segment 1 name]: [brief summary]
- [Segment 2 name]: [brief summary]
- [Segment 3 name]: [brief summary]
How is the voice feeling overall?
- Is the tone consistent and appropriate?
- Are we maintaining the narrative arc?
- Should we adjust anything before continuing?
Continue the iterative process through all sections of the Message Map until complete.
Output: Refined Draft
After all segments are complete:
## Complete Draft
[Full compiled draft]
---
Total length: [word count]
Sections completed: [list all sections]
This draft incorporates:
- Strategic intent from our Discovery phase
- Message structure from our Architecture phase
- Voice calibration from our Framing phase
[If lexicons loaded:]
- Your established narrative patterns and language preferences
Ready for final alignment review?
Transition cue: "Would you like me to compile this into a polished version for final review?"
Goal: Validate resonance and adaptability.
Method: Reflect and refine collaboratively.
Review against Intent Statement:
## Final Alignment Review
Let's verify this draft achieves your original intent:
✓ Purpose: [Original purpose] → Does the draft accomplish this?
✓ Audience: [Target audience] → Will they respond as intended?
✓ Desired Response: [Original goal] → Does the draft evoke this?
✓ Tone: [Original tone objective] → Is the voice appropriate?
✓ Constraints: [Original constraints] → Are these respected?
What feels fully aligned? What needs refinement?
Test the draft's authenticity:
Consider the draft's flexibility:
How might this adapt across contexts?
1. **Platform variations:**
- Email version: [suggest key changes if needed]
- LinkedIn/public version: [suggest adaptations]
- Formal presentation: [note what would change]
2. **Length variations:**
- Condensed (50% length): [identify core elements to keep]
- Expanded (150% length): [suggest where to add depth]
3. **Audience variations:**
- If reader were more senior: [note tone adjustments]
- If reader were less familiar with context: [note explanation needs]
Does the core message maintain coherence across these variations?
Based on alignment review:
Identify any remaining gaps:
Make targeted revisions:
Confirm completion:
Does this draft fully achieve your intent?
Are you ready to move forward with it?
Output: Finalized Message Framework and Draft
## Final Deliverable
[Complete polished draft]
---
### Framework Summary
Intent: [Achieved purpose]
Voice: [Established tone and stance]
Structure: [Narrative arc used]
Key Strengths: [What makes this effective]
[If lexicons used:] Voice Consistency: Aligned with your established patterns
This message is ready for [intended use].
Ask user where to save:
Where should I save this final draft?
Suggestions based on writing type:
[If job-related:]
- Recommended: ~/career-applications/[job-slug]/05-cover-letter-draft.md
- Or: ~/career-applications/[job-slug]/[custom-name].md
[If general professional:]
- Recommended: ~/Documents/[descriptive-name].md
- Or: [user-specified path]
Please provide the full path where you'd like this saved.
Save the file with complete content:
Confirm completion:
✓ Saved to: [file path]
✓ Word count: [count]
✓ Ready for: [next step: review, editing, submission, etc.]
Your collaborative writing process is complete!
Return to previous phases anytime:
Back to Discovery (Phase 1) if:
Back to Architecture (Phase 2) if:
Back to Voice Calibration (Phase 3) if:
Remember: Returning to earlier dialogue is not regression — it's refinement. The Socratic process is iterative, not linear.
This collaborative writing process succeeds when:
Voice authenticity: The final draft sounds genuinely like you
Strategic clarity: The message achieves its intended purpose
Structural integrity: The message is well-organized and flows naturally
Tonal consistency: Voice is appropriate and maintained throughout
User confidence: You feel ready to use this writing
This skill is designed to work excellently in both scenarios:
Both approaches produce high-quality, authentic writing. The difference is whether we're referencing established patterns or discovering them fresh.
skills/career/cover-letter-voice (more structured, job-specific)skills/career/job-description-analysis (understanding the opportunity)skills/career/job-fit-analysis (identifying gaps and strategy)content-media
Download YouTube video transcripts when user provides a YouTube URL or asks to download/get/fetch a transcript from YouTube. Also use when user wants to transcribe or get captions/subtitles from a YouTube video.
development
Transform learning content (like YouTube transcripts, articles, tutorials) into actionable implementation plans using the Ship-Learn-Next framework. Use when user wants to turn advice, lessons, or educational content into concrete action steps, reps, or a learning quest.
tools
Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.
tools
Replace with description of the skill and when Claude should use it.