Skip to content

sync(openapi): agent-sdk/main 69d5249 #316

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
xingyaoww merged 1 commit into from
Feb 10, 2026
Merged

sync(openapi): agent-sdk/main 69d5249 #316

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
122 changes: 119 additions & 3 deletions openapi/agent-sdk.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2462,6 +2462,17 @@
"description": "When to run critic evaluation:\n- 'finish_and_message': Evaluate on FinishAction and agent MessageEvent (default, minimal performance impact)\n- 'all_actions': Evaluate after every agent action (WARNING: significantly slower due to API calls on each action)",
"default": "finish_and_message"
},
"iterative_refinement": {
"anyOf": [
{
"$ref": "#/components/schemas/IterativeRefinementConfig"
},
{
"type": "null"
}
],
"description": "Optional configuration for iterative refinement. When set, Conversation.run() will automatically retry the task if the critic score is below the success_threshold, up to max_iterations."
},
"kind": {
"type": "string",
"const": "APIBasedCritic",
Expand Down Expand Up @@ -2610,6 +2621,17 @@
"description": "When to run critic evaluation:\n- 'finish_and_message': Evaluate on FinishAction and agent MessageEvent (default, minimal performance impact)\n- 'all_actions': Evaluate after every agent action (WARNING: significantly slower due to API calls on each action)",
"default": "finish_and_message"
},
"iterative_refinement": {
"anyOf": [
{
"$ref": "#/components/schemas/IterativeRefinementConfig"
},
{
"type": "null"
}
],
"description": "Optional configuration for iterative refinement. When set, Conversation.run() will automatically retry the task if the critic score is below the success_threshold, up to max_iterations."
},
"kind": {
"type": "string",
"const": "APIBasedCritic",
Expand Down Expand Up @@ -3071,7 +3093,7 @@
"llm"
],
"title": "Agent",
"description": "Main agent implementation for OpenHands.\n\nThe Agent class provides the core functionality for running AI agents that can\ninteract with tools, process messages, and execute actions. It inherits from\nAgentBase and implements the agent execution logic.\n\nExample:\n >>> from openhands.sdk import LLM, Agent, Tool\n >>> llm = LLM(model=\"claude-sonnet-4-20250514\", api_key=SecretStr(\"key\"))\n >>> tools = [Tool(name=\"TerminalTool\"), Tool(name=\"FileEditorTool\")]\n >>> agent = Agent(llm=llm, tools=tools)"
"description": "Main agent implementation for OpenHands.\n\nThe Agent class provides the core functionality for running AI agents that can\ninteract with tools, process messages, and execute actions. It inherits from\nAgentBase and implements the agent execution logic. Critic-related functionality\nis provided by CriticMixin.\n\nExample:\n >>> from openhands.sdk import LLM, Agent, Tool\n >>> llm = LLM(model=\"claude-sonnet-4-20250514\", api_key=SecretStr(\"key\"))\n >>> tools = [Tool(name=\"TerminalTool\"), Tool(name=\"FileEditorTool\")]\n >>> agent = Agent(llm=llm, tools=tools)"
},
"Agent-Output": {
"properties": {
Expand Down Expand Up @@ -3265,7 +3287,7 @@
"kind"
],
"title": "Agent",
"description": "Main agent implementation for OpenHands.\n\nThe Agent class provides the core functionality for running AI agents that can\ninteract with tools, process messages, and execute actions. It inherits from\nAgentBase and implements the agent execution logic.\n\nExample:\n >>> from openhands.sdk import LLM, Agent, Tool\n >>> llm = LLM(model=\"claude-sonnet-4-20250514\", api_key=SecretStr(\"key\"))\n >>> tools = [Tool(name=\"TerminalTool\"), Tool(name=\"FileEditorTool\")]\n >>> agent = Agent(llm=llm, tools=tools)"
"description": "Main agent implementation for OpenHands.\n\nThe Agent class provides the core functionality for running AI agents that can\ninteract with tools, process messages, and execute actions. It inherits from\nAgentBase and implements the agent execution logic. Critic-related functionality\nis provided by CriticMixin.\n\nExample:\n >>> from openhands.sdk import LLM, Agent, Tool\n >>> llm = LLM(model=\"claude-sonnet-4-20250514\", api_key=SecretStr(\"key\"))\n >>> tools = [Tool(name=\"TerminalTool\"), Tool(name=\"FileEditorTool\")]\n >>> agent = Agent(llm=llm, tools=tools)"
},
"AgentBase-Input": {
"$ref": "#/components/schemas/Agent-Input"
Expand Down Expand Up @@ -3485,6 +3507,17 @@
"description": "When to run critic evaluation:\n- 'finish_and_message': Evaluate on FinishAction and agent MessageEvent (default, minimal performance impact)\n- 'all_actions': Evaluate after every agent action (WARNING: significantly slower due to API calls on each action)",
"default": "finish_and_message"
},
"iterative_refinement": {
"anyOf": [
{
"$ref": "#/components/schemas/IterativeRefinementConfig"
},
{
"type": "null"
}
],
"description": "Optional configuration for iterative refinement. When set, Conversation.run() will automatically retry the task if the critic score is below the success_threshold, up to max_iterations."
},
"kind": {
"type": "string",
"const": "AgentFinishedCritic",
Expand All @@ -3507,6 +3540,17 @@
"description": "When to run critic evaluation:\n- 'finish_and_message': Evaluate on FinishAction and agent MessageEvent (default, minimal performance impact)\n- 'all_actions': Evaluate after every agent action (WARNING: significantly slower due to API calls on each action)",
"default": "finish_and_message"
},
"iterative_refinement": {
"anyOf": [
{
"$ref": "#/components/schemas/IterativeRefinementConfig"
},
{
"type": "null"
}
],
"description": "Optional configuration for iterative refinement. When set, Conversation.run() will automatically retry the task if the critic score is below the success_threshold, up to max_iterations."
},
"kind": {
"type": "string",
"const": "AgentFinishedCritic",
Expand Down Expand Up @@ -5516,6 +5560,12 @@
"$ref": "#/components/schemas/SecretRegistry-Output",
"description": "Registry for handling secrets and sensitive data"
},
"agent_state": {
"additionalProperties": true,
"type": "object",
"title": "Agent State",
"description": "Dictionary for agent-specific runtime state that persists across iterations. Agents can store feature-specific state using string keys. To trigger autosave, always reassign: state.agent_state = {**state.agent_state, key: value}. See https://docs.openhands.dev/sdk/guides/convo-persistence#how-state-persistence-works"
},
"title": {
"anyOf": [
{
Expand Down Expand Up @@ -6000,6 +6050,17 @@
"description": "When to run critic evaluation:\n- 'finish_and_message': Evaluate on FinishAction and agent MessageEvent (default, minimal performance impact)\n- 'all_actions': Evaluate after every agent action (WARNING: significantly slower due to API calls on each action)",
"default": "finish_and_message"
},
"iterative_refinement": {
"anyOf": [
{
"$ref": "#/components/schemas/IterativeRefinementConfig"
},
{
"type": "null"
}
],
"description": "Optional configuration for iterative refinement. When set, Conversation.run() will automatically retry the task if the critic score is below the success_threshold, up to max_iterations."
},
"kind": {
"type": "string",
"const": "EmptyPatchCritic",
Expand All @@ -6022,6 +6083,17 @@
"description": "When to run critic evaluation:\n- 'finish_and_message': Evaluate on FinishAction and agent MessageEvent (default, minimal performance impact)\n- 'all_actions': Evaluate after every agent action (WARNING: significantly slower due to API calls on each action)",
"default": "finish_and_message"
},
"iterative_refinement": {
"anyOf": [
{
"$ref": "#/components/schemas/IterativeRefinementConfig"
},
{
"type": "null"
}
],
"description": "Optional configuration for iterative refinement. When set, Conversation.run() will automatically retry the task if the critic score is below the success_threshold, up to max_iterations."
},
"kind": {
"type": "string",
"const": "EmptyPatchCritic",
Expand Down Expand Up @@ -7266,6 +7338,28 @@
"title": "InputMetadata",
"description": "Metadata for task skill inputs."
},
"IterativeRefinementConfig": {
"properties": {
"success_threshold": {
"type": "number",
"maximum": 1.0,
"minimum": 0.0,
"title": "Success Threshold",
"description": "Score threshold (0-1) to consider task successful.",
"default": 0.6
},
"max_iterations": {
"type": "integer",
"minimum": 1.0,
"title": "Max Iterations",
"description": "Maximum number of iterations before giving up.",
"default": 3
}
},
"type": "object",
"title": "IterativeRefinementConfig",
"description": "Configuration for iterative refinement based on critic feedback.\n\nWhen attached to a CriticBase, the Conversation.run() method will\nautomatically retry the task if the critic score is below the threshold.\n\nExample:\n critic = APIBasedCritic(\n server_url=\"...\",\n api_key=\"...\",\n model_name=\"critic\",\n iterative_refinement=IterativeRefinementConfig(\n success_threshold=0.7,\n max_iterations=3,\n ),\n )\n agent = Agent(llm=llm, tools=tools, critic=critic)\n conversation = Conversation(agent=agent, workspace=workspace)\n conversation.send_message(\"Create a calculator module...\")\n conversation.run() # Will automatically retry if critic score < 0.7"
},
"KeywordTrigger": {
"properties": {
"type": {
Expand Down Expand Up @@ -8946,6 +9040,17 @@
"description": "When to run critic evaluation:\n- 'finish_and_message': Evaluate on FinishAction and agent MessageEvent (default, minimal performance impact)\n- 'all_actions': Evaluate after every agent action (WARNING: significantly slower due to API calls on each action)",
"default": "finish_and_message"
},
"iterative_refinement": {
"anyOf": [
{
"$ref": "#/components/schemas/IterativeRefinementConfig"
},
{
"type": "null"
}
],
"description": "Optional configuration for iterative refinement. When set, Conversation.run() will automatically retry the task if the critic score is below the success_threshold, up to max_iterations."
},
"kind": {
"type": "string",
"const": "PassCritic",
Expand All @@ -8968,6 +9073,17 @@
"description": "When to run critic evaluation:\n- 'finish_and_message': Evaluate on FinishAction and agent MessageEvent (default, minimal performance impact)\n- 'all_actions': Evaluate after every agent action (WARNING: significantly slower due to API calls on each action)",
"default": "finish_and_message"
},
"iterative_refinement": {
"anyOf": [
{
"$ref": "#/components/schemas/IterativeRefinementConfig"
},
{
"type": "null"
}
],
"description": "Optional configuration for iterative refinement. When set, Conversation.run() will automatically retry the task if the critic score is below the success_threshold, up to max_iterations."
},
"kind": {
"type": "string",
"const": "PassCritic",
Expand Down Expand Up @@ -9928,7 +10044,7 @@
"version": {
"type": "string",
"title": "Version",
"default": "1.11.1"
"default": "1.11.2"
},
"docs": {
"type": "string",
Expand Down