agentleFS
Sign inStart free
glifxyzglifxyz/glif-mcp-server/skills/get-started/SKILL.md

get-started

Skill213 starsChanged 37 days agoMITView on GitHub

Help someone start with Glif. Explain how it connects, check the connection and credits for free, and suggest what to make first. Use when the user asks what Glif is, how to set it up, whether it's connected, which account is connected, how many credits they have, or what Glif can do. Never starts generation.

23 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 22 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Check the connection. If Glif's tools are available, call whoami.

    line 9
  2. Distinctive:

    Never ask for passwords, API keys or sign-in codes in chat.

    line 12
  3. Very specific:

    If the user asks which account or how many credits, report only the whoami fields they asked about, such as username, available credits or …

    line 14

“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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Help the user get oriented. The user's explicit instructions come first, within host safety rules.

  1. Check the connection. If Glif's tools are available, call whoami; it's free and proves the connection works. If they're missing, help the user connect:
    • Claude Code: claude plugin marketplace add glifxyz/glif-mcp-server, then claude plugin install glif@glif, then /mcp to sign in.
    • Any other client: add the remote MCP server https://glif.app/api/mcp and sign in when the browser opens. https://glif.app/mcp has steps for each client. Never ask for passwords, API keys or sign-in codes in chat. A free account can be created at https://glif.app.
  2. Explain what costs credits. Only compose_project spends the account's Glif credits. Connecting, browsing, choosing a reference and checking status are free. Video uses more credits than images. Glif decides the real cost when it runs. Plans and credits are managed at https://glif.app/pricing; never buy or upgrade in chat.
  3. If the user asks which account or how many credits, report only the whoami fields they asked about, such as username, available credits or recent spend.
  4. To browse, call list_projects, then get_project for the one they pick. Show media links from the result. If the host renders Glif's viewer, view_media can reopen saved media in it.
  5. Explain how continuity works:
    • A project keeps its messages and media. Passing its project_id to compose_project adds new work to the same project.
    • A new reference file goes through upload_file. Its returned fileUrl goes in compose_project's attachments.
    • Project IDs, job IDs, asset IDs and media URLs are different values. Do not swap them.
  6. End with a choice that fits what the user is doing. Examples:
    • In a codebase: "add a hero image and texture to this page", "make a logo and favicon set", "record a 30-second demo of this app", "make App Store screenshots", "add UI sounds".
    • Anywhere: "make a YouTube thumbnail", "remove the background from this image", "create a character sheet", "make an explainer video", "generate a voiceover", "plan an ad campaign", "show my recent projects". Or show a draft brief for review.

Onboarding never calls compose_project, uploads files, or changes settings. Browsing, picking a reference or accepting a suggestion is not a request to generate. Show draft prompts as text and wait for an explicit "go".

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free