Start recording screen

Stop recording screen and save to local

Ensure every discovered skill has an initial DB record. For each skill in *discovered_skills* (``SkillMeta`` objects from :meth:`Skil

Apply an edit with retry on failure. If the first attempt fails (patch parse error, path mismatch, etc.), feeds the error back to the

Rank candidates using embedding cosine similarity.

Execute a list of evolution contexts in parallel (throttled). Used by all three triggers after building/confirming contexts.

Extract package name from uvx arguments. Args: args: uvx arguments list, e.g. ["--from", "office-powerpoint-mcp-server",

Extract package name from uv run arguments. Args: args: uv arguments list, e.g. ["run", "--with", "biomcp-python", "biomc

Format recent analyses into a concise context block for prompts.

Get a OpenSpaceClient instance (raises CloudError if not configured).

Get or create aiohttp session.

Check if the collected traceback is a harmless error.

Fallback cleanup for platforms where process-tree cleanup is unavailable.

Load .env files once per process. Search order (first-loaded wins for each key): 1. ``openspace/.env`` (package root — works regardless of

Load single JSON configuration file. This function wraps the generic load_json_file and adds global configuration specific error handling and

Load and parse nanobot config.json. Returns None on failure.

Re-quote SKILL.md frontmatter in-place so values are valid YAML.

Parse a SKILL.md file into a SkillMeta. Only ``name`` and ``description`` are read from frontmatter (per the official skill format).

Parse tool data into Tool objects.

Rebuild text-to-key mapping for quick lookup.

Find the OpenClaw config file on disk.

Run evolution as a token-driven agent loop. Modeled after ``GroundingAgent.process()`` — the loop continues until the LLM outputs an

Run a command via asyncio subprocess and return a result dict matching the format returned by the local_server endpoints.

Return a collision-safe prefix for all cache entries of a skill.

Stable short ID for a workflow directory, unique across roots. Uses a hash suffix derived from the resolved path to avoid collisions when dir

Analyze agent actions and generate statistics.

Resolve grounding config: inline JSON > file path > None. Supports: * ``OPENSPACE_CONFIG_JSON`` — inline JSON string (written to a temp fi

Capture screenshot via HTTP API. Calls: GET {base_url}/screenshot Returns: PNG image bytes, or

return True if allowed, False if denied. Command check uses token-level matching to prevent simple space/escape bypasses.

Close an MCP session by session name. Args: session_name: Session name in format 'mcp-<server_name>'

Collect all text files in a skill directory. Returns ``{relative_path: content}``. Binary files are silently skipped.

Disconnect from GUI desktop environment

Scan all skill_dirs and populate the registry. Each skill is a sub-directory containing a ``SKILL.md`` file. The ``skill_id`` is read

Internal helper. Guarantee that `initialize()` has been executed

Execute a task with OpenSpace. Args: task: Task instruction context: Additional context. Communicati

Execute a Python command on desktop environment. Used for pyautogui commands. Args: command: Python comm

* Fuzzy search: check if query characters appear in target string in order

Get configuration for specified backend

Get base URL for connecting to local server. Priority: 1. Environment variable LOCAL_SERVER_URL 2. Read from local_server/config

Read local server configuration. Priority: 1. Environment variable LOCAL_SERVER_URL (parsed into host/port) 2. Config file local

Get the current mcp-use package version.

Get tools with low success rate (candidates for review/removal). Args: success_rate_threshold: Tools below this

Get current desktop screenshot.

Get comprehensive system information. Returns information including: - system: OS name (Linux, Darwin, Windows)

Negotiate with the backend, discover tools, etc. Return session information (can be an empty dict). `self.tools` nee

Check if the server configuration is for a stdio server. Args: server_config: The server configuration section Returns: True

Load configuration files

Return the SKILL.md content (with frontmatter stripped) for *skill_id*.

Parse YAML frontmatter into a flat dict. Simple line-by-line parser (no PyYAML dependency). Handles both quoted and unquoted values. Retu

Read the OpenSpace env block from the current host agent config. Resolution order: 1. nanobot — ``tools.mcpServers.openspace.env`` 2.