Skip to content
Agentmart
Submit a tool

export

Load for /vellum:export.

Skill by FiredMosquito831

IndexedApache-2.0★ 2
Write a review

Chalk Labs

Need help deploying export?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/firedmosquito831-export

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Load for /vellum:export. The export skill is the operator's manual for building the final manuscript: it runs the readiness gate, interprets a failure as a prioritized fix list, runs the deterministic build on PASS, and explains the manifest and the pandoc fallback. The engine implementation lives in scripts/vellum_lib/export.py (Workstream B); this skill is what the operator (muse) follows.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/export

What's inside

Skill name
export
Description
Load for /vellum:export. The export skill is the operator's manual for building the final manuscript: it runs the readiness gate, interprets a failure as a prioritized fix list, runs the deterministic build on PASS, and explains the manifest and the pandoc fallback. The engine implementation lives in scripts/vellum_lib/export.py (Workstream B); this skill is what the operator (muse) follows.
License
Apache-2.0
Instructions
44 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • files: referenced but missing: scripts/vellum_lib/export.py

Files (3)

  • SKILL.md
  • assembly.md · resources/
  • manifest.md · resources/

SKILL.md (first lines)

# Export

The export skill is the operator's manual for `/vellum:export` — the command that builds the final manuscript artifact. It is **not** the build engine (that is `scripts/vellum_lib/export.py`, Workstream B); it is the sequence the operator follows, the gate it runs first, and the way it interprets success and failure. The manuscript on disk is untouched — export is a build artifact, assembled from the gated sources.

## The export sequence

1. **Run `vellum readiness`.** This evaluates the preconditions (`gates/resources/readiness.md`): every chapter `final` with complete frontmatter, ledgers clean, the readiness report PASSES, the word-budget report attached, pivotal chapters have blind artifacts.
2. **If PASS** → run `vellum export build --out export/manuscript-<date> [--epub]`. Interpret the manifest (`resources/manifest.md`).
3. **If not PASS** → readiness prints a prioritized missing list. Present it to the author; do not run the build.

Muse does not re-litigate the gate. Muse runs `vellum readiness`, reads its output, and either runs the build or presents the fix list. The gate is the engine's verdict.

## Interpreting a failure

A failed readiness check prints a prioritized fix list. The format mirrors the beta-reader REVISE output: the changes that would most raise the verdict, each with location, evidence, intervention kind, and the axis/gate it moves. The author works the list, re-runs `/vellum:export`, and the gate re-evaluates. Common failures:

- **Incomplete frontmatter** on a chapter → `vellum ledger check` emits `frontmatter:incomplete`; the author (or `@kb-lead`) fills the fields.
- **Open ledger findings** → resolve or dismiss-with-author-signoff (`vellum dismiss`).
- **Readiness report REVISE** → the beta reader's prioritized fix list; revise and re-run the beta read.
- **Pivotal chapter missing blind artifact** → run the blind read for that chapter.
- **Cold-read BLOCKER** → resolve or triage.

## On PASS — the build

`vellum export build --out export/manuscript-<date>` assembles (`resources/assembly.md`):

- Title page, copyright placeholder, front/back matter, joined chapters (in frontmatter `number` order), and the manifest.
- `--epub` adds a stdlib-zipfile EPUB with a stable `urn:uuid:<book-uuid>` identifier from `kb/story.md` so highlights survive rebuilds. **No pandoc required** — the EPUB writer is stdlib zipfile.
- DOCX/PDF via pandoc when present; otherwise the markdown bundle + instructions. Pandoc is optional, never on the critical path.

## The manifest

The build writes `manifest.md` (`resources/manifest.md`): chapter list, word counts, sha256 checksums, and **gate provenance** for every chapter (outline-approval date, blind verdict, beta scores, logged overrides). Export never silently drops a gate — the provenance is attached. The manifest is the build's certificate; it is not the manuscript.

## Load the resource needed

- `resources/assembly.md` — the deterministic assembly phase: order of parts, chapter-joining conventions, scene-break rules, the output files, and gate provenance.
- `resources/manifest.md` — the manifest fields: build metadata, chapter list, per-chapter gate provenance, book-level gates, overrides, and checksums.

## What export is not

Export does not edit the manuscript, does not re-litigate the gates, and does not replace the readiness report or the blind/beta artifacts — it references them. The manuscript is the source of truth; the export is a build artifact assembled from it. If the manuscript changes after export, the author re-runs the build (the manifest's checksums detect the drift).

From FiredMosquito831/my-writing-companion, Apache-2.0.

Signals

Stars
2
Forks
0
Open issues
0
Last commit (this folder)
Sep 9, 2026 (1 month ago)
Last release
Sep 11, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from FiredMosquito831

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get export running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying export?

CallGet help
Cite this page

export skill. Agent Mart. https://agentmart.dev/skills/firedmosquito831-export (last updated 2026-10-24)