Return a logger with *name* (defaults to ``openspace``). The first call triggers :meth:`configure` automatically.

Execute tool and auto-capture screenshot.

Priority: user_value → config → default

Load cache from disk. Returns empty dict if not exists.

Open a temporary read-only connection. WAL mode allows concurrent readers and one writer. Each read operation gets its own connection

Run the full search pipeline on candidates. Each candidate dict should have at minimum: - ``skill_id``, ``name``, ``description``

Percentage savings: (a - b) / a * 100. Positive = saved.

Single-round LLM call with optional tool execution. Args: messages: conversation history (List[Dict] for standar

Send a JSON-RPC request and return the result. Args: method: The JSON-RPC method name (e.g., "tools/list", "tools/call")

Get global configuration instance. Usage: - Get configuration in Provider: get_config().get_backend_config('shell') - Ge

Exit the async context manager. Args: exc_type: The exception type, if an exception was raised. exc_val: The exceptio

Record one step operation Args: backend: backend type (gui/shell/mcp/web/system) tool: tool name (na

Save tool metadata to disk (overwrites existing cache). Args: servers: Dict mapping server_name -> list of tool

Close the persistent connection. Subsequent ops will raise. Performs a WAL checkpoint before closing so that all committed data is fl

List all discovered skills.

Ensure the connector is connected.

Delete all quality data.

Unified diff (git diff format) between two strings.

Flush any queued telemetry events

Get trajectory directory path

Collect all text files in a directory (recursive). Excludes the ``.skill_id`` sidecar (internal metadata, not skill content).

Ensure the connector is connected, reconnecting if necessary. Raises: RuntimeError: If connection cannot be established and auto_

Add log message

Bind runtime information to the tool instance. Allow the tool to be invoked directly without specifying backend/session/server.

Close all sessions and release resources. Automatically called when using context manager.

List tools from backend(s) or session. 1. session_name is provided → return tools from that session 2. backend is li

Load agent actions from a trajectory directory.

Load a single :class:`SkillRecord` by id.

Reset call source to its previous value.

Write to stderr, filtering out harmless error messages. Args: s: The string to write Returns:

Lazy-initialise the OpenSpace engine.

Safely parse tool_call.function.arguments which may be JSON string. Handles: 1. Proper JSON strings with true/false/null 2. P

Parse a connector result dict into ``(stdout, stderr, returncode)``.

Execute HTTP request. Returns ``(status_code, response_body)``.

Send a request and wait for a response.

Deserialize a skill_records row + related rows → SkillRecord.

* Escape HTML to prevent XSS

Flush any remaining buffered content and the underlying stream.

Load skill records, keyed by ``skill_id``. Args: active_only: If True, only return records with ``is_active=True``.

Load complete recording session including trajectory, metadata, plans, and snapshots. Args: recording_dir: Path to recording dir