From 626labs
Use this skill when the user mentions writing, publishing, theses, blog posts, essays, fiction, audiobook pipelines, ThesisStudio, BlogStudio, WriterStudio, or anything that lives under 626Labs-Publishing. Trigger on phrases like "new article", "starting an essay", "where does this draft live", "lift to packages", "library snapshot". Orients you to the studios/works/library/packages bucket model and routes you to the right Lead Writer persona instead of drafting prose at the umbrella level.
How this skill is triggered — by the user, by Claude, or both
Slash command
/626labs:publishingThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
The umbrella for all 626 Labs writing — academic, blog, fiction, finished outputs. **Not a git repo.** Each studio, each active work, and (eventually) each shared package is its own independent git repo. The umbrella `CLAUDE.md` at `C:\Users\estev\Projects\626Labs-Publishing\CLAUDE.md` is the orchestrator-level keystone — read it first when invoked.
The umbrella for all 626 Labs writing — academic, blog, fiction, finished outputs. Not a git repo. Each studio, each active work, and (eventually) each shared package is its own independent git repo. The umbrella CLAUDE.md at C:\Users\estev\Projects\626Labs-Publishing\CLAUDE.md is the orchestrator-level keystone — read it first when invoked.
| Bucket | Contains | Mutability |
|---|---|---|
studios/ | Reusable templates, one per genre (placeholders only — [YOUR TITLE], [CHARACTER]) | Active — improved over time, never holds finished prose |
works/ | Real writing in progress, authored against a studio | Active — current authoring lives here |
library/ | Finished, published works as frozen snapshots | Frozen — represents a moment of publication |
packages/ | Cross-studio shared skills and pipelines | Active — lift point for anything two-or-more studios share |
The placement rule: active work → works/, finished work → library/, reusable scaffolding → studios/, shared code → packages/.
| Studio | Purpose | Path | Dashboard project |
|---|---|---|---|
| ThesisStudio | Academic theses, dissertations, master's papers, long-form research articles | studios/ThesisStudio/ | CooIu5ieQ5QXzczcQ0HL |
| BlogStudio | Blog posts, voice-driven short-form writing | studios/BlogStudio/ | FLfcTlVUOcURie1HuXYT |
| WriterStudio | Fiction — novellas, short stories, serial chapters | studios/WriterStudio/ | BLbTIFk07cTSLFOqomEJ |
Each studio has its own CLAUDE.md with a Lead Writer persona, three pillars, and reference architecture. Defer to those when working inside a studio — the umbrella does not override studio-level personas.
These are pipelines two or more studios will share. They don't exist as their own repos yet — tracked here so the future lift target is clear.
| Package | Source studio | Targets | Status |
|---|---|---|---|
audiobook-pipeline (ElevenLabs speaker tagging → audiobook) | WriterStudio | All three studios | To-extract |
lay-translator (academic-register → plain-language) | ThesisStudio | BlogStudio (today), WriterStudio (planned) | To-generalize |
Do not lift prematurely. Extract a package when a second studio actually needs it; shared design intent without shared code is a documentation problem, not a packaging problem.
At the umbrella (626Labs-Publishing/ itself):
packages/, how studios relate to works to library assets.Inside a studio (studios/<X>/):
CLAUDE.md is authoritative. This skill is background context.Inside a work (works/<X>/):
CLAUDE.md (it inherits from the studio) for project-specific conventions, voice, and scope.packages/.package.json / tokens.yaml.Tag dashboard decisions to the right project ID:
Mdy4yqKsrnaubrxJXM6b.Use mcp__626Labs__manage_decisions with action: log.
Ask: "What are you working on — a new piece, an existing work, lifting something to packages, or library archival?" Then route to the right studio (and its Lead Writer persona) rather than answering at the umbrella level.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub estevanhernandez-stack-ed/626labs-plugin --plugin 626labs