agentleFS
Sign inStart free
glifxyzglifxyz/glif-mcp-server/skills/continue-project/SKILL.md

continue-project

Skill213 starsChanged 37 days agoMITView on GitHub

Find the user's saved Glif project, show its media, and add new work to it without starting a separate project. Use when: "show my recent Glif projects", "open my last image", "keep going on that one", "make 5 variations to A/B test", "same but vertical", "make a version for TikTok", "resize it to every format", "same character in a new scene", "fix only the text", "fix scene 3", "next episode", "do another one like yesterday's".

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    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. Distinctive:

    Use this when the user wants to revisit or build on saved Glif work.

    line 8
  2. Very specific:

    Read it with get_project. Set include_messages only when you need the history to understand the request.

    line 13
  3. Very specific:

    If get_project shows an active job, the project is already working.

    line 15
  4. Distinctive:

    If the user only asked to prepare a continuation, show the draft prompt and wait.

    line 21

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Use this when the user wants to revisit or build on saved Glif work. The user's explicit instructions come first, within host safety rules.

  1. Find the project:
    • From a link or ID the user gives you.
    • Or with list_projects, paging with cursor only as far as needed. Read it with get_project. Set include_messages only when you need the history to understand the request. If several projects match, ask. If a project is missing or inaccessible, say so; do not guess other IDs or storage URLs.
  2. For a browse-only request, show the media links from get_project and stop. If the host renders Glif's viewer, view_media with the project_id (and optional asset_ids) reopens it there. Browsing never starts generation.
  3. If get_project shows an active job, the project is already working. Do not start another. Follow it with get_job_status only if this conversation started that job; otherwise share the project link.
  4. Work out what stays the same and what changes. Name the chosen media in the brief and keep the project's project_id. Do not re-upload existing media just to copy it. Common follow-ups, each phrased as one clear change:
    • Variations: "Make 4 variations of image 2 that change only the background." Keep the count small unless the user asks for more.
    • New format: "Same video, recomposed for 9:16 without cropping the subject."
    • Targeted fix: "Fix only the headline text to read '...'. Keep everything else identical."
    • Continuity: "Same character and style, new scene: ..." or "Next episode, same format and voice."
  5. If the user only asked to prepare a continuation, show the draft prompt and wait. Choosing media is not approval to spend.
  6. For a clear request to generate, note that it uses the account's Glif credits. Call compose_project with the project_id and the brief. Keep related outputs in one call. Leave intelligence unset so the project keeps its tier, unless the user asks to change it.
  7. Share the projectUrl. Call get_job_status with the jobId as job_id until the job finishes, and show new media as it arrives without repeats. Report the result or the failure. Never retry paid work on your own.

New work is added to the project. Originals stay untouched. Do not change account settings, buy credits or set up automatic generation while continuing a project.

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