Exporting and sharing

Every product has two ways in, and they want different shapes:

A skill upload takes one skill at a time. Its archive must hold exactly one SKILL.md, with the skill folder as the archive's top level and no plugin manifest.
A plugin may hold many skills, but needs its manifest and skills/<name>/.

Uploading the wrong shape fails with "Zip must contain exactly one SKILL.md file".

Which export goes where

Where it goes	Command	What you get
Claude or ChatGPT skill settings (upload)	`skillgendev export --one-skill --zip`	`dist/skill/<name>.zip`: `<name>/SKILL.md` plus a file per page. One skill, no manifest
Claude Code or Cowork on your machine	`skillgendev export`	`skills/<page>/` folders to copy into `~/.claude/skills/`. One skill per page
Claude Code plugin	`skillgendev export --plugin --zip`	`.claude-plugin/plugin.json` + `skills/<page>/`
ChatGPT and Codex plugin	`skillgendev export-chatgpt --zip`	`plugin.json` + one skill under `skills/<name>/` + `assets/icon.svg`

export --one-skill and export-chatgpt both bundle, so both hold exactly one SKILL.md. A plain export --zip of several pages does not, and says so when you run it.

Every export copies all four fill methods, so the same skill works in Claude, Cowork, ChatGPT or any other AI.

Before anything leaves your machine

Every export, of every shape:

Checks each skill against the Agent Skills spec and skillgendev's own rules.
Scans every file for secrets and personal data: private keys, AWS, GitHub, Slack, Anthropic or OpenAI and Google keys, JSON web tokens, bearer tokens, password assignments, real-looking email addresses, phone numbers and source file paths. Reserved example domains (example.com, .test, localhost and similar) and 555-01xx numbers are allowed, because generated examples use them on purpose. Any finding stops the export.
Writes MANIFEST.sha256 with a SHA-256 hash of every file, sorted by path, so a tampered skill folder can be detected.

Internal data never goes: page.json, evidence, probe captures and questions stay in .skillgendev/. evals/ is left out of bundles, because it is for you, not for the AI.

One skill, a file per page

--one-skill keeps every page whole and makes the entry point cheap to read:

acme/
  SKILL.md                                  # the only SKILL.md: a table of forms, ~25 lines
  references/acme-campaign-new/page.md            # that page's fields, rules and fill call
  references/acme-campaign-new/locators.md        # and its other fill methods
  references/acme-campaign-new/manual-steps.md
  references/acme-campaign-new/fields.md, layout.json
  scripts/acme-campaign-new/fill.js               # its fallback engine
  MANIFEST.sha256

The entry point lists each form's title, route, what is on it and the one file to read, then says to read exactly one. A fill costs that table plus one page file, so adding pages does not make a fill dearer. Paths inside each page's files are rewritten to match this layout.

The folder is named after the skill (--name, else the app prefix), because every product requires the folder name and the name in SKILL.md to match. With --zip that holds whatever you call the file: --out dist/download still zips as acme/SKILL.md.

Options

Flag	`export`	`export-chatgpt`
`--zip`	Write one zip file	Write one zip, ready to upload
`--name`	Plugin or bundled skill name (default: the app's prefix)	Plugin and skill name (default: the skill prefix)
`--out`	Destination (default `dist/<name>`; `dist/skill/<name>` with `--one-skill`)	Destination (default `dist/chatgpt/<name>`)
`--one-skill`	Bundle every page into one skill	always bundled
`--plugin`	Lay out as a Claude Code plugin	n/a

Both plugin manifests name the author chapeee, the maintainer's recorded decision; no flag changes it.

The export stops and says so when --name is too long: a skill name is at most 64 characters, and ChatGPT counts <plugin>:<skill> against that limit.

ChatGPT and Codex

Skill upload: skillgendev export --one-skill --zip, then upload it in ChatGPT's skill settings: the same file Claude takes.
Plugin: skillgendev export-chatgpt --zip writes the portable Agent Plugins layout that ChatGPT and Codex share: plugin.json at the root (name, version, description, author, and the listing name, tagline and icon), one skill under skills/<name>/, and assets/icon.svg. Upload it at platform.openai.com/plugins as a skills-only plugin; OpenAI scans and reviews it before you choose to publish.

To use it yourself, copy the folder to ~/.codex/plugins/acme, list it in ~/.agents/plugins/marketplace.json, restart the ChatGPT desktop app and install it from the Plugins Directory:

{
  "name": "local",
  "plugins": [
    {
      "name": "acme",
      "source": { "source": "local", "path": "./.codex/plugins/acme" },
      "policy": { "installation": "AVAILABLE", "authentication": "ON_INSTALL" },
      "category": "Productivity"
    }
  ]
}

The skill index

skillgendev index

writes skills-index.html and skills-index.json at the repository root: one self-contained page with an interactive graph of the app (app, skills, sections, fields and their dependencies) and a card per skill. It needs no network, so it works opened from a zip. export includes it.

Which export goes where​

Before anything leaves your machine​

One skill, a file per page​

Options​

ChatGPT and Codex​

The skill index​