# AgentScope

## Docs

- [What's AgentScope 2.0?](https://docs.agentscope.io/versions/2.0.7dev/en/index.md): Secure, efficient, flexible, and complete.
- [Quickstart](https://docs.agentscope.io/versions/2.0.7dev/en/quickstart.md): Get up and running with AgentScope 2.0 in minutes
- [Release Notes](https://docs.agentscope.io/versions/2.0.7dev/en/release-notes.md): Version history of AgentScope 2.x, including new features, changes, and bug fixes grouped by module for each release.
- [Message & Event](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/message-and-event.md): The core data abstractions for agent communication and streaming
- [Overview](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/agent/overview.md): The stateless reasoning-acting loop engine at the core of AgentScope
- [Configure Agent](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/agent/configure-agent.md): Assemble an agent from a model, tools, and configuration
- [Run Agent](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/agent/run-agent.md): Run agents and work with their replies, context, and state
- [Interrupt Agent](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/agent/interrupt-agent.md): Stop a running or paused agent cleanly and resume from a consistent state
- [Human-in-the-Loop](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/agent/human-in-the-loop.md): Pause for user confirmation or external execution, then resume with result events
- [Overview](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/model/overview.md): Connect model providers with a credential and discover their models through model cards
- [LLM](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/model/llm.md): Create chat models, call them, and plug in your own provider
- [TTS](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/model/tts.md): Turn text into speech, in standard or realtime streaming mode
- [Embedding](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/model/embedding.md): Turn text and media into vectors for search, RAG, and memory
- [Overview](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/context/overview.md): Manage the agent's working memory to keep long tasks on track
- [Environment Awareness](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/context/environment-awareness.md): Keep the agent aware of time, tasks, and context usage as they change
- [Compress Context](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/context/compress-context.md): Keep the context length within the preset limit
- [Offload Context](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/context/offload-context.md): Persist dropped content so the agent can read it back on demand
- [Overview](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/tool/overview.md): Equip agents with tools, MCP servers, and skills through the toolkit
- [Python Tool](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/tool/python-tool.md): Build tools from classes or plain functions, and control how they run
- [MCP](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/tool/mcp.md): Connect agents to MCP servers and use their tools
- [Skill](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/tool/skill.md): Extend agent capabilities with markdown instruction sets
- [Meta Tool](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/tool/manage-tools.md): Let the agent activate and deactivate tool groups at runtime
- [Plan](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/plan.md): Give agents a structured task list to plan, track, and coordinate complex work
- [Overview](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/permission-system/overview.md): How rules, modes, and tool-level checks decide every tool call
- [Permission Mode](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/permission-system/permission-mode.md): Pick the global policy that fits how your agent is deployed
- [Permission Rule](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/permission-system/permission-rule.md): Write allow, deny, and ask patterns for specific tools and calls
- [Tool-Level Checks](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/permission-system/tool-check.md): Runtime safety analysis each tool performs on its own inputs
- [Middleware](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/middleware.md): Intercept and extend agent behavior at key lifecycle points
- [RAG](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/rag.md): Build retrieval-augmented generation (RAG) capabilities for agents.
- [Long-Term Memory](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/long-term-memory.md): Cross-session long-term memory implemented with agent middleware
- [Overview](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/workspace/overview.md): Give the agent an execution environment it can act in and persist to
- [Run Workspace](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/workspace/run-workspace.md): Create a workspace on any backend and wire it into an agent
- [Manage Resources](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/workspace/manage-resources.md): Add and remove MCP servers and skills at runtime
- [MCP Gateway](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/workspace/mcp-gateway.md): How sandboxed workspaces expose their MCP servers to the host
- [Console](https://docs.agentscope.io/versions/2.0.7dev/en/building-blocks/console.md): Quickly test and verify agent behavior in the terminal
- [Architecture](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/agent-service.md): Host your agent as a multi-tenant, multi-session HTTP service
- [Agent Team](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/agent-team.md): Leader agents that spawn and coordinate worker agents through built-in team tools
- [Resource Sharing](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/sharing.md): Share API credentials, agents, and knowledge bases as a team or organization
- [RAG Service](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/rag.md): A one-click multi-tenant, distributed RAG service
- [Workspace Manager](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/workspace-manager.md): Manage the lifecycle of agent harness workspace
- [Overview](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/hub/overview.md): Plug sources of MCP servers and skills into your service.
- [MCP Hub](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/hub/mcp-hub.md): Let users install MCP servers from a registry with their own credentials.
- [Skill Hub](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/hub/skill-hub.md): Let users install skills published by others and equip their agents with them.
- [Overview](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/channel/overview.md): Bring the agents in your AgentScope service onto IM platforms.
- [Message routing](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/channel/routing.md): Decide which agent answers an incoming message and which session it joins.
- [Feishu](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/channel/feishu.md): Chat with your service's agents inside Feishu.
- [Discord](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/channel/discord.md): Chat with your service's agents inside Discord.
- [Custom channel](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/channel/custom.md): Implement ChannelBase to connect a platform beyond the built-ins.
- [[Deprecated] Legacy sectioned schema — use /schema/v2](https://docs.agentscope.io/api-reference/agent/[deprecated]-legacy-sectioned-schema-—-use-schemav2.md): Return the legacy sectioned JSON Schema fragments.
- [Full AgentData JSON Schema for the agent form](https://docs.agentscope.io/api-reference/agent/full-agentdata-json-schema-for-the-agent-form.md): Return the full :class:`AgentData` JSON Schema.
- [List all agents](https://docs.agentscope.io/api-reference/agent/list-all-agents.md): Return all agent records belonging to the authenticated user.
- [Create a new agent](https://docs.agentscope.io/api-reference/agent/create-a-new-agent.md): Create and persist a new agent configuration.
- [Delete an agent](https://docs.agentscope.io/api-reference/agent/delete-an-agent.md): Permanently delete an agent configuration.
- [Update an agent](https://docs.agentscope.io/api-reference/agent/update-an-agent.md): Partially update an existing agent configuration.
- [Trigger a chat run (fire-and-forget)](https://docs.agentscope.io/api-reference/chat/trigger-a-chat-run-fire-and-forget.md): Trigger a chat run for the specified session.
- [List JSON schemas for all credential types](https://docs.agentscope.io/api-reference/credential/list-json-schemas-for-all-credential-types.md): Return JSON schemas for all registered credential types.
- [List all credentials](https://docs.agentscope.io/api-reference/credential/list-all-credentials.md): Return all credential records belonging to the authenticated user.
- [Create a new credential](https://docs.agentscope.io/api-reference/credential/create-a-new-credential.md): Store a new credential.
- [Delete a credential](https://docs.agentscope.io/api-reference/credential/delete-a-credential.md): Permanently delete a credential.
- [Update a credential](https://docs.agentscope.io/api-reference/credential/update-a-credential.md): Replace the payload of an existing credential.
- [Report service liveness and per-component readiness](https://docs.agentscope.io/api-reference/health/report-service-liveness-and-per-component-readiness.md): Report whether the service is ready to serve requests.
- [List Mcp Hubs](https://docs.agentscope.io/api-reference/hub/list-mcp-hubs.md): Return every registered MCP hub.
- [List Mcp Cards](https://docs.agentscope.io/api-reference/hub/list-mcp-cards.md): Browse or search one MCP hub's catalog.
- [Get Mcp Card](https://docs.agentscope.io/api-reference/hub/get-mcp-card.md): Return one MCP card, including the inputs the user must fill.
- [Install Mcp](https://docs.agentscope.io/api-reference/hub/install-mcp.md): Fill a card's template and add the result to the user's library.
- [List Skill Hubs](https://docs.agentscope.io/api-reference/hub/list-skill-hubs.md): Return every registered skill hub.
- [List Skill Cards](https://docs.agentscope.io/api-reference/hub/list-skill-cards.md): Browse or search one skill hub's catalog.
- [Get Skill Card](https://docs.agentscope.io/api-reference/hub/get-skill-card.md): Return one skill card, including its ``SKILL.md`` body.
- [Install Skill](https://docs.agentscope.io/api-reference/hub/install-skill.md): Add a skill card to the user's library.
- [List embedding models compatible with the KB dimension policy](https://docs.agentscope.io/api-reference/knowledge_bases/list-embedding-models-compatible-with-the-kb-dimension-policy.md): List embedding models the user can pick at KB-creation time.
- [JSON Schema for the KB middleware's tunable parameters](https://docs.agentscope.io/api-reference/knowledge_bases/json-schema-for-the-kb-middlewares-tunable-parameters.md): Return the parameter schema for :class:`agentscope.middleware.RAGMiddleware`.
- [List file types the configured parsers can ingest](https://docs.agentscope.io/api-reference/knowledge_bases/list-file-types-the-configured-parsers-can-ingest.md): Advertise the union of media types and filename extensions every registered parser accepts.
- [List the caller's knowledge bases](https://docs.agentscope.io/api-reference/knowledge_bases/list-the-callers-knowledge-bases.md): Return all knowledge bases owned by the authenticated user.
- [Create a new knowledge base](https://docs.agentscope.io/api-reference/knowledge_bases/create-a-new-knowledge-base.md): Create a new knowledge base for the authenticated user.
- [Delete a knowledge base](https://docs.agentscope.io/api-reference/knowledge_bases/delete-a-knowledge-base.md): Permanently delete a knowledge base.
- [Update mutable fields on a knowledge base](https://docs.agentscope.io/api-reference/knowledge_bases/update-mutable-fields-on-a-knowledge-base.md): Update mutable fields on a knowledge base.
- [List documents registered in a knowledge base](https://docs.agentscope.io/api-reference/knowledge_bases/list-documents-registered-in-a-knowledge-base.md): List every document registered against a knowledge base.
- [Upload a document into a knowledge base](https://docs.agentscope.io/api-reference/knowledge_bases/upload-a-document-into-a-knowledge-base.md): Register an uploaded document and dispatch it for indexing.
- [Batch-query indexing status of one or more documents](https://docs.agentscope.io/api-reference/knowledge_bases/batch-query-indexing-status-of-one-or-more-documents.md): Return the current lifecycle state of a batch of documents.
- [Delete a document from a knowledge base](https://docs.agentscope.io/api-reference/knowledge_bases/delete-a-document-from-a-knowledge-base.md): Remove a document and all its chunks from a knowledge base.
- [Search a knowledge base by natural-language query](https://docs.agentscope.io/api-reference/knowledge_bases/search-a-knowledge-base-by-natural-language-query.md): Run a similarity search over a knowledge base.
- [List Mcps](https://docs.agentscope.io/api-reference/mcp/list-mcps.md): Return every MCP the user has installed, ordered by name.
- [Delete Mcp](https://docs.agentscope.io/api-reference/mcp/delete-mcp.md): Remove an MCP from the user's library.
- [Update Mcp](https://docs.agentscope.io/api-reference/mcp/update-mcp.md): Rename, enable/disable, or re-key an installed MCP.
- [List all schedules](https://docs.agentscope.io/api-reference/schedule/list-all-schedules.md): List all schedules owned by the current user.
- [Create a new schedule](https://docs.agentscope.io/api-reference/schedule/create-a-new-schedule.md): Create a new schedule and register it with the scheduler.
- [Delete a schedule](https://docs.agentscope.io/api-reference/schedule/delete-a-schedule.md): Permanently delete a schedule.
- [Update a schedule](https://docs.agentscope.io/api-reference/schedule/update-a-schedule.md): Partially update a schedule.
- [List execution sessions for a schedule](https://docs.agentscope.io/api-reference/schedule/list-execution-sessions-for-a-schedule.md): Return all sessions triggered by a given schedule.
- [List sessions for an agent](https://docs.agentscope.io/api-reference/sessions/list-sessions-for-an-agent.md): Return all sessions for an agent as enriched :class:`SessionView` entries.
- [Create a new session](https://docs.agentscope.io/api-reference/sessions/create-a-new-session.md): Create (or resume) a session for a given agent and workspace.
- [Delete a session](https://docs.agentscope.io/api-reference/sessions/delete-a-session.md): Permanently delete a session and all its associated state.
- [Update a session](https://docs.agentscope.io/api-reference/sessions/update-a-session.md): Update the model configuration of an existing session.
- [Interrupt a running or HITL-parked chat run for a session](https://docs.agentscope.io/api-reference/sessions/interrupt-a-running-or-hitl-parked-chat-run-for-a-session.md): Request interruption of an in-progress reply for a session.
- [List messages for a session](https://docs.agentscope.io/api-reference/sessions/list-messages-for-a-session.md): Return persisted messages for a session.
- [Probe the session's high-level status](https://docs.agentscope.io/api-reference/sessions/probe-the-sessions-high-level-status.md): Return the unified :class:`SessionStatus` for a session.
- [Subscribe to a session's event stream (SSE)](https://docs.agentscope.io/api-reference/sessions/subscribe-to-a-sessions-event-stream-sse.md): Subscribe to a session's live event stream.
- [List Skills](https://docs.agentscope.io/api-reference/skill/list-skills.md): Return every skill the user has installed, ordered by name.
- [Get Skill](https://docs.agentscope.io/api-reference/skill/get-skill.md): Return one installed skill, including its ``SKILL.md`` body.
- [Delete Skill](https://docs.agentscope.io/api-reference/skill/delete-skill.md): Remove a skill from the user's library.
- [List Mcps](https://docs.agentscope.io/api-reference/workspace/list-mcps.md): Return all MCP clients with live tool list and health status.
- [Add Mcp](https://docs.agentscope.io/api-reference/workspace/add-mcp.md): Add an MCP client to the session's workspace.
- [Add Mcps From Library](https://docs.agentscope.io/api-reference/workspace/add-mcps-from-library.md): Put MCPs the user has already installed into this workspace.
- [Remove Mcp](https://docs.agentscope.io/api-reference/workspace/remove-mcp.md): Remove an MCP client from the session's workspace by name.
- [List Skills](https://docs.agentscope.io/api-reference/workspace/list-skills.md): Return all skills available in the session's workspace.
- [Add Skill](https://docs.agentscope.io/api-reference/workspace/add-skill.md): Add a skill to the session's workspace from the given path.
- [Upload Skill](https://docs.agentscope.io/api-reference/workspace/upload-skill.md): Install a skill from an uploaded folder.
- [Add Skills From Library](https://docs.agentscope.io/api-reference/workspace/add-skills-from-library.md): Put skills the user has already installed into this workspace.
- [Remove Skill](https://docs.agentscope.io/api-reference/workspace/remove-skill.md): Remove a skill from the session's workspace by name.
- [List Workspace Directory](https://docs.agentscope.io/api-reference/workspace/list-workspace-directory.md): List one directory level, reachable from a session's workspace.
- [Create Download Token](https://docs.agentscope.io/api-reference/workspace/create-download-token.md): Mint a short-lived token for a browser-native download.
- [Read Workspace File](https://docs.agentscope.io/api-reference/workspace/read-workspace-file.md): Stream one file out of a session's workspace.
- [List all candidate models under the given credential type](https://docs.agentscope.io/api-reference/model/list-all-candidate-models-under-the-given-credential-type.md): Return all candidate models under the given credential type.
- [List all candidate TTS models under the given credential type](https://docs.agentscope.io/api-reference/tts-model/list-all-candidate-tts-models-under-the-given-credential-type.md): Return all candidate TTS models under the given credential type.
- [List all candidate embedding models under the given credential type](https://docs.agentscope.io/api-reference/embedding-model/list-all-candidate-embedding-models-under-the-given-credential-type.md): Return all candidate embedding models under the credential type.
- [List Channel Types](https://docs.agentscope.io/api-reference/channels/list-channel-types.md): List supported channel types with their JSON schemas.
- [List Channels](https://docs.agentscope.io/api-reference/channels/list-channels.md): List channels owned by the current user.
- [Create Channel](https://docs.agentscope.io/api-reference/channels/create-channel.md): Create a channel.
- [Get Channel](https://docs.agentscope.io/api-reference/channels/get-channel.md): Get channel details.
- [Delete Channel](https://docs.agentscope.io/api-reference/channels/delete-channel.md): Delete a channel.
- [Update Channel](https://docs.agentscope.io/api-reference/channels/update-channel.md): Update routing / session / config / enabled.
- [Enable Channel](https://docs.agentscope.io/api-reference/channels/enable-channel.md): Enable a channel.
- [Disable Channel](https://docs.agentscope.io/api-reference/channels/disable-channel.md): Disable a channel.
- [Channel Status](https://docs.agentscope.io/api-reference/channels/channel-status.md): The channel's live connection status.
- [List Channel Sessions](https://docs.agentscope.io/api-reference/channels/list-channel-sessions.md): Sessions this channel spawned, newest first.
- [List Chat Ids](https://docs.agentscope.io/api-reference/channels/list-chat-ids.md): Known chats for routing config: platform list ∪ passively seen.
- [FAQ](https://docs.agentscope.io/versions/2.0.7dev/en/others/faq.md): Frequently asked questions about AgentScope v2.0
- [Changelog](https://docs.agentscope.io/versions/2.0.7dev/en/others/change-log.md): Core differences between AgentScope 2.0 and 1.0
- [What's ReMe?](https://docs.agentscope.io/reme/latest/en/overview.md): Learn how ReMe turns conversations and resources into user-owned, file-native agent memory.
- [Quick Start](https://docs.agentscope.io/reme/latest/en/quickstart.md): Install ReMe and complete a verifiable write, index, search, and read loop.
- [Concepts](https://docs.agentscope.io/reme/latest/en/concepts.md): The principles and trade-offs behind ReMe's file-native agent memory.
- [Overview](https://docs.agentscope.io/reme/latest/en/integration/overview.md): Choose how an application or agent should run and use ReMe.
- [Deployment](https://docs.agentscope.io/reme/latest/en/integration/deployment.md): Run ReMe as a shared service and choose MCP, CLI, or HTTP as the access boundary.
- [Python SDK](https://docs.agentscope.io/reme/latest/en/integration/python-sdk.md): Run ReMe in the host Python process and manage its lifecycle directly.
- [Agent Integrations](https://docs.agentscope.io/reme/latest/en/integration/plugins.md): Connect supported agents to ReMe through maintained in-process Python, tool, Skill, and plugin integrations.
- [Reference](https://docs.agentscope.io/reme/latest/en/reference.md): Stable workspace, configuration, CLI, HTTP, and MCP contracts for ReMe.
- [FAQ](https://docs.agentscope.io/reme/latest/en/faq.md): Answers and troubleshooting steps for common ReMe problems.
- [Contribution](https://docs.agentscope.io/reme/latest/en/contribution.md): Set up ReMe development, run tests, and prepare a pull request.

## OpenAPI Specs

- [openapi](https://docs.agentscope.io/versions/2.0.7dev/en/deploy/openapi.json)

## Optional

- [Home Page](https://agentscope.io/)
- [Blog](https://agentscope.io/blog/)