MCPcopy Create free account

hub / github.com/HKUDS/OpenSpace / functions

Functions2,195 in github.com/HKUDS/OpenSpace

↓ 1 callersFunction_normalize_legacy_keys
(raw: Dict[str, Any])
openspace/communication/config.py:326
↓ 1 callersMethod_normalize_visibility_value
Treat legacy/group-shared non-public skills as private locally.
openspace/cloud/client.py:105
↓ 1 callersMethod_normalize_webhook_payload
(self, payload: dict[str, Any])
openspace/communication/adapters/feishu.py:330
↓ 1 callersMethod_normalize_websocket_event
(self, data: Any)
openspace/communication/adapters/feishu.py:362
↓ 1 callersFunction_normalize_whatsapp_identifier
(value: Any)
openspace/communication/adapters/whatsapp.py:459
↓ 1 callersMethod_open_control_socket
(self)
openspace/communication/adapters/whatsapp.py:181
↓ 1 callersFunction_parse_add_file_content
Collect ``+``-prefixed lines for an Add File hunk.
openspace/skill_engine/patch.py:567
↓ 1 callersMethod_parse_analysis
Convert the raw LLM JSON output into an ExecutionAnalysis. Also attaches observed tool execution records from ``traj.jsonl`` so the a
openspace/skill_engine/analyzer.py:841
↓ 1 callersMethod_parse_computer_tool_use
Parse Anthropic computer tool use to pyautogui command. Args: tool_input: Tool input from Anthropic (action, coo
openspace/grounding/backends/gui/anthropic_client.py:442
↓ 1 callersMethod_parse_confirmation
Parse LLM confirmation response (expects JSON with 'proceed' field).
openspace/skill_engine/evolver.py:668
↓ 1 callersMethod_parse_evolution_output
Extract edit content or failure reason from LLM output. MUST only be called when ``EVOLUTION_COMPLETE`` or ``EVOLUTION_FAILED`` is pr
openspace/skill_engine/evolver.py:1242
↓ 1 callersFunction_parse_host_from_env
(default: str = "127.0.0.1")
openspace/mcp_server.py:936
↓ 1 callersMethod_parse_llm_response
Parse LLM response to extract action. Args: response: LLM response (should be JSON) Returns:
openspace/grounding/backends/gui/tool.py:448
↓ 1 callersFunction_parse_patch_header
Parse a ``*** Add/Delete/Update File:`` header line.
openspace/skill_engine/patch.py:540
↓ 1 callersFunction_parse_port_from_env
(default: int = 8080)
openspace/mcp_server.py:922
↓ 1 callersMethod_parse_skill_selection_response
Parse the LLM response and extract selected skill IDs + plan. Returns: (skill_ids, brief_plan)
openspace/skill_engine/registry.py:707
↓ 1 callersFunction_parse_update_chunks
Parse ``@@``-delimited change chunks for an Update File hunk.
openspace/skill_engine/patch.py:582
↓ 1 callersMethod_passes_prefilter
(self, message: ChannelMessage)
openspace/communication/adapters/feishu.py:478
↓ 1 callersFunction_pick_first_env
(names: tuple[str, ...])
openspace/host_detection/resolver.py:70
↓ 1 callersMethod_plan_next_action
Use LLM/VLM to plan the next action. This method sends: - Task description - Current screenshot (vision inpu
openspace/grounding/backends/gui/tool.py:265
↓ 1 callersFunction_port_flag_was_set
(argv: list[str])
openspace/mcp_server.py:919
↓ 1 callersMethod_prefilter_skills
Narrow the candidate set using BM25 + embedding hybrid ranking. Keeps at most ``max(15, max_skills * 5)`` candidates for the LLM sele
openspace/skill_engine/registry.py:505
↓ 1 callersFunction_prepare_tools_for_llmclient
Convert BaseTool list to LLMClient usable format, with deduplication. Args: tools: BaseTool instance list (should be obtained from Gr
openspace/llm/client.py:95
↓ 1 callersFunction_print_status
Print system status
openspace/__main__.py:112
↓ 1 callersFunction_print_task_summary
(record: Dict)
gdpval_bench/run_benchmark.py:745
↓ 1 callersMethod_process_message
(self, session: ChannelSession, message: ChannelMessage)
openspace/communication/gateway.py:374
↓ 1 callersMethod_process_tool_degradation_locked
Inner body of process_tool_degradation, called under _degradation_lock.
openspace/skill_engine/evolver.py:332
↓ 1 callersFunction_read_or_create_skill_id
Read ``skill_id`` from ``.skill_id`` sidecar, or create one. The sidecar file is a single-line plain-text file containing only the ``skill_id
openspace/skill_engine/registry.py:44
↓ 1 callersFunction_read_upload_meta
Read upload metadata with three-tier fallback. Resolution order: 1. ``.upload_meta.json`` sidecar file (written right after evolution)
openspace/mcp_server.py:295
↓ 1 callersMethod_receive_loop
(self, ws: aiohttp.ClientWebSocketResponse)
openspace/communication/adapters/whatsapp.py:205
↓ 1 callersMethod_receive_messages
Continuously receive and process messages from the WebSocket.
openspace/grounding/backends/mcp/transport/connectors/websocket.py:78
↓ 1 callersMethod_record_agent_execution
Record agent execution to recording manager. Args: result: Execution result instruction: Original in
openspace/agents/grounding_agent.py:926
↓ 1 callersMethod_record_gui
(self, tool_call, result)
openspace/recording/manager.py:743
↓ 1 callersFunction_record_matches_live_process
(record: dict[str, Any])
openspace/communication/gateway_runtime.py:90
↓ 1 callersMethod_record_mcp
(self, tool_call, result, server: str)
openspace/recording/manager.py:715
↓ 1 callersMethod_record_shell
(self, tool_call, result)
openspace/recording/manager.py:804
↓ 1 callersMethod_record_system
(self, tool_call, result)
openspace/recording/manager.py:866
↓ 1 callersMethod_record_to_quality_manager
Record execution result to quality manager for quality tracking.
openspace/grounding/core/tool/base.py:286
↓ 1 callersMethod_record_tool_quality_feedback
Feed LLM-identified tool issues to the ToolQualityManager. **Deduplication**: The rule-based system already records each tool call as
openspace/skill_engine/analyzer.py:275
↓ 1 callersMethod_record_web
(self, tool_call, result)
openspace/recording/manager.py:899
↓ 1 callersMethod_register_providers_from_config
Based on GroundingConfig.enabled_backends, register Provider instances to self._registry. Here only do *instantiation*, not a
openspace/grounding/core/grounding_client.py:59
↓ 1 callersMethod_register_self
Register the agent class in the registry upon instantiation.
openspace/agents/base.py:149
↓ 1 callersMethod_register_system_provider
Register SystemProvider separately because it requires GroundingClient instance. SystemProvider provides meta-level tools for queryin
openspace/grounding/core/grounding_client.py:102
↓ 1 callersMethod_register_temp_file
Register a temporary file for later cleanup
openspace/local_server/health_checker.py:69
↓ 1 callersMethod_remember_sent_message_id
(self, message_id: str)
openspace/communication/adapters/feishu.py:501
↓ 1 callersMethod_render_agents
Render agents section
openspace/utils/ui.py:250
↓ 1 callersMethod_render_grounding
Render grounding operations section
openspace/utils/ui.py:282
↓ 1 callersMethod_render_header
Render header section
openspace/utils/ui.py:221
↓ 1 callersMethod_render_logs
Render logs section
openspace/utils/ui.py:337
↓ 1 callersMethod_reorder_servers
Reorder servers dict according to _server_order.
openspace/grounding/backends/mcp/tool_cache.py:34
↓ 1 callersMethod_resize_screenshot
Resize screenshot to display size for Computer Use API. For computer-use-2025-01-24, the screenshot must be resized to the
openspace/grounding/backends/gui/anthropic_client.py:146
↓ 1 callersMethod_resolve_bridge_script
(self)
openspace/communication/adapters/whatsapp.py:384
↓ 1 callersFunction_resolve_config_path
(path: Optional[str])
openspace/communication/config.py:223
↓ 1 callersFunction_resolve_mentions
(text: str, mentions: list[Any])
openspace/communication/adapters/feishu.py:776
↓ 1 callersMethod_resolve_path
Resolve relative paths against the task workspace (data_path).
openspace/grounding/backends/shell/productivity_tools.py:272
↓ 1 callersFunction_resolve_references
Resolve reference file paths relative to GDPVal dataset dir.
gdpval_bench/task_loader.py:388
↓ 1 callersFunction_resolve_tool_call_target
Resolve a returned tool name to a concrete tool object. The LLM is expected to return the deduped tool key from ``tool_map``. Some providers
openspace/llm/client.py:189
↓ 1 callersFunction_resolve_transport
(requested_transport: str, argv: list[str])
openspace/mcp_server.py:939
↓ 1 callersMethod_restore_logs
(self)
openspace/__main__.py:59
↓ 1 callersMethod_restore_mcp_logging
Restore MCP SDK logging to normal.
openspace/grounding/backends/mcp/transport/task_managers/stdio.py:342
↓ 1 callersMethod_rewrite_nonleading_system_messages_for_minimax
Rewrite non-leading system messages into internal user notes for MiniMax.
openspace/llm/client.py:470
↓ 1 callersMethod_rollback_start
(self, exc: Exception)
openspace/communication/gateway.py:230
↓ 1 callersMethod_run
(self, **kwargs)
openspace/grounding/core/tool/local_tool.py:11
↓ 1 callersMethod_run_analysis_loop
Run analysis as an agent loop with optional tool use. Most analyses complete in a single pass (LLM outputs JSON directly). When the t
openspace/skill_engine/analyzer.py:710
↓ 1 callersFunction_run_gateway
(config_path: Optional[str])
openspace/communication/gateway.py:534
↓ 1 callersFunction_run_phase_concurrent
Run one phase with N concurrent OpenSpace workers. Trade-offs vs serial: - Skills from concurrent tasks may not be visible to each other (r
gdpval_bench/run_benchmark.py:613
↓ 1 callersFunction_run_phase_serial
Run one phase sequentially. A single OpenSpace instance is reused so that skills accumulate within the phase.
gdpval_bench/run_benchmark.py:567
↓ 1 callersMethod_run_shell_command
Run a shell command string (used for conda-wrapped scripts).
openspace/grounding/backends/shell/transport/local_connector.py:295
↓ 1 callersMethod_sanitize_and_save_cache
Sanitize raw cache and save to sanitized cache file.
openspace/grounding/backends/mcp/provider.py:273
↓ 1 callersFunction_sanitize_schema
Sanitize tool parameter schema to comply with Claude API requirements. Fixes common issues: - Empty object schemas (no properties, no req
openspace/llm/client.py:23
↓ 1 callersMethod_save_all_sync
Synchronous full save (used by async wrapper and migration).
openspace/grounding/core/quality/store.py:211
↓ 1 callersMethod_save_tools_to_cache
Save tools metadata to cache file.
openspace/grounding/backends/mcp/provider.py:401
↓ 1 callersMethod_scale_coordinates
Scale coordinates from display size to actual screen size. The API returns coordinates in display_size (e.g., 1024x768).
openspace/grounding/backends/gui/anthropic_client.py:160
↓ 1 callersFunction_scan_workflow_tree
(directory: Path, discovered: Dict[str, Path], *, _depth: int = 0, _max_depth: int = 6)
openspace/dashboard_server.py:443
↓ 1 callersMethod_scan_workspace_files
Scan workspace directory and collect file information. Args: workspace_path: Path to workspace directory
openspace/agents/grounding_agent.py:590
↓ 1 callersFunction_schema_to_openai
Convert ToolSchema to OpenAI ChatCompletion tool format
openspace/llm/client.py:72
↓ 1 callersFunction_scope_hash
(identity: str)
openspace/communication/gateway_runtime.py:23
↓ 1 callersMethod_score_phase
Compute hybrid score = vector_score + lexical_boost.
openspace/cloud/search.py:168
↓ 1 callersMethod_select_and_inject_skills
Select skills for task via LLM, inject into GroundingAgent. When the registry has many skills, a BM25 + embedding pre-filter narrows
openspace/tool_layer.py:689
↓ 1 callersMethod_select_key_screenshots
Select key screenshots from a sequence, preferring first/last/evenly-spaced.
openspace/agents/visual_analyzer.py:213
↓ 1 callersMethod_serialize_atspi_element
Serialize AT-SPI element to dictionary Args: element: AT-SPI accessible element depth: Current depth
openspace/local_server/platform_adapters/linux_adapter.py:164
↓ 1 callersMethod_serialize_ax_element
Serialize macOS accessibility element to dictionary Args: element: AX element depth: Current depth
openspace/local_server/platform_adapters/macos_adapter.py:218
↓ 1 callersMethod_serialize_response_field
Convert provider response fields into plain Python containers.
openspace/llm/client.py:514
↓ 1 callersMethod_serialize_uia_element
Serialize Windows UIA element to dictionary Args: element: UIA element depth: Current depth
openspace/local_server/platform_adapters/windows_adapter.py:231
↓ 1 callersMethod_session_usage_bytes
(self, session_key: str)
openspace/communication/attachment_cache.py:108
↓ 1 callersMethod_session_worker
( self, session: ChannelSession, queue: asyncio.Queue[ChannelMessage], )
openspace/communication/gateway.py:322
↓ 1 callersMethod_set_embedding
Store embedding in structured cache.
openspace/grounding/core/search_tools.py:324
↓ 1 callersFunction_set_global_state_for_productivity
Set ClawWork global state so productivity tools have data_path/date.
openspace/grounding/backends/shell/productivity_tools.py:44
↓ 1 callersMethod_setup_security_policy
(self, config: dict | None = None)
openspace/grounding/core/provider.py:30
↓ 1 callersFunction_setup_ui
(args)
openspace/__main__.py:431
↓ 1 callersMethod_show_cursor
Show cursor
openspace/utils/ui.py:105
↓ 1 callersFunction_snapshot_skills
Dump all active skills from the SkillStore.
gdpval_bench/run_benchmark.py:170
↓ 1 callersFunction_split_text
(content: str, limit: int)
openspace/communication/adapters/feishu.py:794
↓ 1 callersFunction_split_text
(content: str, limit: int)
openspace/communication/adapters/whatsapp.py:424
↓ 1 callersMethod_start_bridge_process
(self)
openspace/communication/adapters/whatsapp.py:342
↓ 1 callersMethod_start_websocket_client
(self)
openspace/communication/adapters/feishu.py:267
↓ 1 callersMethod_stdout_supports_color
()
openspace/utils/logging.py:295
↓ 1 callersMethod_stop_session_workers
(self)
openspace/communication/gateway.py:241
↓ 1 callersFunction_stratified_sample
Pick N tasks per occupation for balanced coverage. Groups tasks by ``occupation``, takes the first ``per_occupation`` from each group, and re
gdpval_bench/task_loader.py:364
↓ 1 callersMethod_subtree
( self, conn: sqlite3.Connection, sid: str, depth: int, visited: set,
openspace/skill_engine/store.py:1153
← previousnext →801–900 of 2,195, ranked by callers