Skip to content

Conversation

@xiaow190
Copy link
Collaborator

  • Support model-predicted actions: add_object, remove_object, position
  • Parse model output with ... tag
  • Dispatch to corresponding handlers (add/remove/position) based on action type
  • Update flagscale scene info accordingly

xiaow190 and others added 30 commits July 8, 2025 17:28
- Fixed import path case issue for `Collaborator` to ensure cross-platform compatibility.
- Replaced editable Git dependency with pinned `mcp==1.10.1` for stable installation.
…s API

- Replaced stdio-based MCP connection with HTTP-based streamable client using config-defined URL.
- Updated `config.yaml` to include MCP server URL under `mcp.URL`.
- Refactored `robot_status` endpoint to return structured list of registered robot states.
- Removed unused Flask imports and simplified code structure.
- Fixed inconsistent indentation in the Docker run command
- Added missing container path for the `-v` volume mount
- Removed trailing whitespace after `-p 8888:8888`
- Improved readability with consistent parameter alignment
…ass for better modularity and maintainability
…h scene info

- Implemented read_all_environment to fetch and parse all environment data from Redis
- Updated MASTER_PLANNING_PLANNING prompt to include scene_info and require task decomposition to consider available objects in the scene
- document required fields in master/slaver config.yaml
  including model.cloud_server, collaborator, profile, tool, and robot
- provide examples for slaver.robot in both local and remote call modes
- clarify default values and usage to help users correctly configure RoboOS
@xiaow190 xiaow190 merged commit ee1e0d4 into FlagOpen:stand-alone Aug 28, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant