draft
Skill2 starsChanged 53 days agoMITView on GitHub
Turn an interview transcript into an anchor piece in the user's codified voice, then gate it through the six-persona Writers Council at a hard 9.0 threshold. Offers multiple hooks and closings. The transcript's words are the only allowed source material. Trigger phrases include "draft it", "write the piece", "/draft".
24 other projects use a version of this.
Adopted by
Most recently taken up by
- Mharbulous/probanda-devtheir own version, first seen Nov 8, 2026
- PeterPartridge/CooperativeAICodingtheir own version, first seen Nov 8, 2026
- yuryalencar/ai-skillstheir own version, first seen Nov 1, 2026
- jigripokri/POHAtheir own version, first seen Nov 1, 2026
- danicat/speedgraphertheir own version, first seen Oct 17, 2026
- choiyounggi/auto-velogtheir own version, first seen Oct 15, 2026
Used with
Projects with it also use onboard, status and 2 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
SentrySkillonboard
Onboard a service to sentry-options. Use when integrating a new service, creating schemas, or setting up the multi-repo sentry-options workflow.
Also in 3 projects that use this
AnthropicSkillstatus
Case status summary by audience — client-facing (plain language), internal (for the professor), or court-ready (formal caption format per local rules). Same facts, different framing and depth. Use when a student needs to update the client, brief the professor, or prepare a court status report.
Also in 3 projects that use this
VercelSkillsetup
Set up Vercel CLI and project configuration. Use when the user says "set up Vercel", "configure Vercel", "link to Vercel", or "vercel init".
Also in 4 projects that use this
SentrySkillreview
Fixture review skill.
Also in 5 projects that use this
Other versions24 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 24
- anthropics/claude-for-legal★ 9.5ksame name
- jigripokri/POHA★ 121same name
- tance-mang/chinese-webnovel-skills★ 97same name
- danicat/speedgrapher★ 52same name
- richard-kim-79/archora-skills★ 49same name
- drafthq/draft★ 40same name
- alirezarezvani/gaios★ 37same name
- coolxeo/grok-bot-skills★ 7same name
- vercel-labs/workflow-campaign-demos★ 6same name
- iwe-org/marketing-workspace★ 5same name
14 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- content-machineSkill
- content-setupSkill
- interviewSkill
- lessonsSkill
- oracleSkill
- repurposeSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 0 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 22 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 10
Read first, all four: the interview transcript (from /interview), and the personal folder's profile.md, voice-guide.md, content-lessons.md.
- Very specific: line 22
Ask which anchor format (from profile.md's anchor list): long LinkedIn, newsletter section, essay, thread script.
- Very specific: line 26
Check every item in content-lessons.md — these are past mistakes.
- Very specific: line 33
Before the council: list any sentence not traceable to the transcript (excluding hook/closing/transitions).
“Common” is measured: kept by at least 6 in 10 of the 22 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Draft — shape the clay, never invent it
Read first, all four: the interview transcript (from /interview), and the personal folder's profile.md, voice-guide.md, content-lessons.md. Then reference/council-personas.md. Personal folder resolution: ~/.content-machine/redirect.md first line if present, else ~/.content-machine/.
The transcript-words rule (the machine's core gate)
Every substantive sentence in the draft must be traceable to the interview transcript. You may reorder, trim, splice, and fix grammar. You may not add claims, examples, framings, or vocabulary the user didn't produce. Exceptions: the hook and the closing line, plus connective tissue (transitions ≤ ~6 words). The writer's job is shaping clay — the user already made the clay.
❌ Violation: transcript says "he rebuilt it in a week"; draft says "he rebuilt it in a week — proof that the era of the mega-consultancy is ending." (second clause invented) ✅ Compliant: transcript says "McKinsey said six months. He did it in a week with Claude Code. Nobody at the bank has looked at consultants the same way since." → draft uses exactly these sentences, tightened.
Steps
1. Confirm format
Ask which anchor format (from profile.md's anchor list): long LinkedIn, newsletter section, essay, thread script. One format per run.
2. Draft
- Structure and rhythm come from
voice-guide.md(their hook formulas, their structures, their #1 rule). - Check every item in
content-lessons.md— these are past mistakes; repeating a logged lesson is the one unforgivable error. - Body: transcript words only, per the rule above.
3. Hooks and closings
Offer 3 hook options and 2 closing options (the one place invention is allowed), each in a distinct register from the voice guide's hook formulas. Label which formula each uses.
4. Self-check the transcript-words rule
Before the council: list any sentence not traceable to the transcript (excluding hook/closing/transitions). If any exist, rewrite or delete them. State "transcript check: clean" or the violations found.
5. Writers Council
Run per reference/council-personas.md: six independent scores 1–10 with one-line justifications and one fix each. Aggregate mean ≥ 9.0 to pass. On failure: fix only the two lowest dimensions, re-run all six. Max 3 loops, then present best-with-scores. If the personal sources.md names a local slop-gate script, run it during the Allergist's turn; a failing exit caps the Allergist at 4.
6. Present
Show: final draft, the hook/closing options, the council scorecard, and the loop count. Save to drafts/YYYY-MM-DD-<spike-slug>.md in the personal folder with the transcript path in frontmatter.
What happens next
User edits/approves → /lessons captures what the machine got wrong → /repurpose fans out derivatives.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

