| name | content-start |
| description | The front door to the Content pack: turn one idea into a week of content. Trigger on "turn this idea into content", "one idea into a week of content", "help me with content", "make content from this", "plan my content", "I have a topic, help me post it", "what should I post", or any open-ended content request where the user has something to say but has not picked a format. You bring one idea; the OS checks whether it knows your voice, then routes you to the outcome you want: a single post, the same idea across every channel, a sequenced campaign, or a reply to something incoming. One honest disclaimer, no full onboarding, and it never publishes for you. Routes to linkedin-post, content-repurposer, campaign-from-theme, review-responder, voice-interview, or brand-voice-interview depending on the outcome chosen.
|
| why | A founder doing their own marketing has the idea but not the patience to learn six content skills. This is the one entry that turns one idea into the format they actually need and makes sure it sounds like them, not like generic AI. |
| enhance | Set your voice profile first (say 'set up my voice profile'). With it, every draft sounds like you wrote it; without it, the writing falls back to safe anti-AI defaults that read clean but generic. |
| summary | Bring one idea; the OS makes it sound like you and aims it at the channel you want. |
| allowed-tools | ["Read","Bash"] |
| mcp_requirements | [] |
Content Start
Runs on: reasoning - this skill reads whether the OS knows your voice and routes you to the writing skill that produces the content. It does not write the content itself and it never publishes. On a read-only or cloud surface, it explains the same route without claiming a draft was saved.
This is the front door to content. You have one idea: a lesson, a story, an opinion, a customer message that needs answering. Underneath that one idea sit several outcomes: a single post, that idea stretched across LinkedIn, Instagram, YouTube and email, a full campaign, or a reply to something a customer sent you. This skill makes sure the OS knows your voice, then aims the idea at the outcome you want.
The principle: do more for you than you asked, and never publish a word in a voice that is not yours. If the OS does not yet hold your voice, it says so and offers to capture it before it drafts, because a draft in the wrong voice is worse than no draft.
The outcomes behind one idea
| You want | What it means | Where it routes |
|---|
| One post | a single strong post in your voice | linkedin-post |
| A week of it | the same idea adapted across your channels | content-repurposer |
| A campaign | a sequenced, funnel-aware run of posts toward one goal | campaign-from-theme |
| A reply | answer an incoming review, DM, or message in the right voice | review-responder |
Not sure which? Say "just make the post" and the OS runs linkedin-post first (the cheapest, most immediately useful), then names what repurposing and a full campaign would add, so you choose with a real draft in front of you.
Read the state before routing
The route is only as good as what it knows. Before choosing, read brain/.snapshot.md. If it is missing, or its date: line is more than 3 days old, run python scripts/brain-snapshot.py --write first and read the fresh one. The open-flags block tells you what is stuck, the must-do block tells you what this week is actually about, and the patterns block tells you what keeps happening - a front door that routes without them is a cold start at the exact moment the user is deciding whether this OS knows them. If Python is unavailable, route from the files you can read and say so.
The flow
