10 actions in 8 pieces of software · ranked by words, meaning and measured trust · 631 ms

1. 1Meta: Muse Spark 1.3 Contributorverified and signedpassport

   modelrecord

   • tool-calling

     Can call tools (function calling).

   • text+image+video+file->text

     Takes text, image, video, file; returns text.

   • structured-outputs

     Can return JSON matching a schema.

2. 2Meta: Muse Spark Openverified and signedpassport

   modelrecord

   • text+image+video+file+audio->text

     Takes text, image, video, file, audio; returns text.

3. 3Muse-Spark-Open-816B-A42Bverified and signedpassport

   modelrecord1,809 downloads/month

   • image-text-to-text

     Task: image-text-to-text

4. 4Muse Linkfound in code, not yet checkedno passport yet

   MCP servergitWRLD999/muse-link

   • agent_tools_status

     Report actual installation/readiness of winapp, UFO app APIs, CPU OmniParser, free SideCursor and MouseMux. CUA Driver remains the default scoped desktop input route.

5. 5PR Sparkverified and signedpassport

   libraryws2git/pr-sparkrecord

   • ws2git/pr-spark

     Automatically create pull requests between branches using GitHub’s API for workflows and automation.

6. 6sparkverified and signedpassport

   skillwishworldbetter/seedex-skillsrecord

   • spark

     Use when the user wants to brainstorm an idea or design a feature/spec. Explores intent and requirements through dialogue, then writes a spec document to docs/spark/ and STOPS. Does not auto-chain to implementation planning or any other skill.

7. 7Muse-Chat-MCPfound in code, not yet checkedno passport yet

   MCP serverduclm1x1/Muse-Chat-MCP

   • muse_login

     Open muse.ai in the dedicated Chrome window and wait for Meta sign-in to complete. Use when muse_status reports loggedIn=false.

8. 8muse-image-mcpfound in code, not yet checkedno passport yet
   • edit_image

     以 Meta Muse 模型依既有圖片與指令生成新圖。images 可填本機檔案路徑或 http(s) 網址，