# FIM One

> AI-Powered Connector Hub — embed into one system as a Copilot, or connect them all as a Hub.

## Docs

- [Get Agent](https://docs.fim.ai/api-reference/agents/get-agent.md)
- [List Agents](https://docs.fim.ai/api-reference/agents/list-agents.md)
- [Create Conversation](https://docs.fim.ai/api-reference/conversations/create-conversation.md)
- [Delete Conversation](https://docs.fim.ai/api-reference/conversations/delete-conversation.md)
- [Get Conversation](https://docs.fim.ai/api-reference/conversations/get-conversation.md)
- [List Conversations](https://docs.fim.ai/api-reference/conversations/list-conversations.md)
- [Update Conversation](https://docs.fim.ai/api-reference/conversations/update-conversation.md)
- [Dag Endpoint](https://docs.fim.ai/api-reference/dag-endpoint.md): Run a DAG planner pipeline with SSE progress updates.
- [Inject Message](https://docs.fim.ai/api-reference/inject-message.md): Inject a user message into a running agent execution.
- [List Kbs](https://docs.fim.ai/api-reference/knowledge-bases/list-kbs.md)
- [Retrieve From Kb](https://docs.fim.ai/api-reference/knowledge-bases/retrieve-from-kb.md)
- [React Endpoint](https://docs.fim.ai/api-reference/react-endpoint.md): Run a ReAct agent query with SSE progress updates.
- [Create My Api Key](https://docs.fim.ai/api-reference/user-settings/create-my-api-key.md): Create a new personal API key. The full key is returned only once.
- [Delete My Api Key](https://docs.fim.ai/api-reference/user-settings/delete-my-api-key.md): Permanently delete one of the user's own API keys.
- [List My Api Keys](https://docs.fim.ai/api-reference/user-settings/list-my-api-keys.md): List the authenticated user's own API keys (paginated).
- [Toggle My Api Key Active](https://docs.fim.ai/api-reference/user-settings/toggle-my-api-key-active.md): Enable or disable one of the user's own API keys.
- [Authentication](https://docs.fim.ai/api/authentication.md): How to authenticate with the FIM One API
- [API Overview](https://docs.fim.ai/api/overview.md): Getting started with the FIM One REST API
- [Agent & Resource Discovery](https://docs.fim.ai/architecture/agent-discovery.md): How FIM One discovers and assembles resources — the dual-mode architecture of agent-constrained and global auto-discovery.
- [Connector Architecture](https://docs.fim.ai/architecture/connector-architecture.md): How FIM One connects legacy systems through AI — from Copilot to Hub.
- [Context Management](https://docs.fim.ai/architecture/context-management.md): How FIM One manages conversation context — a five-layer defense-in-depth system that prevents token overflow while preserving conversation quality.
- [DAG Engine](https://docs.fim.ai/architecture/dag-engine.md): How FIM One's DAG engine works — dynamic planning, parallel execution, goal-aware analysis, and autonomous re-planning.
- [Design Philosophy](https://docs.fim.ai/architecture/design-philosophy.md): Architectural decisions made during DAG execution design — what we chose NOT to build, and why.
- [Hook System](https://docs.fim.ai/architecture/hook-system.md): Deterministic enforcement that sits below the LLM loop and cannot be bypassed by agent instructions.
- [LLM Provider Compatibility](https://docs.fim.ai/architecture/llm-provider-guide.md): How FIM One routes LLM calls, the tool_choice architecture, and provider-specific pitfalls — especially Anthropic thinking + AWS Bedrock.
- [Market](https://docs.fim.ai/architecture/market.md): How the Market organizes and distributes resources — from two-tier classification to unified subscription.
- [Organizations](https://docs.fim.ai/architecture/organization.md): Team-level resource sharing with role-based access, optional review, and credential fallback.
- [Philosophy](https://docs.fim.ai/architecture/philosophy.md): Why dynamic planning is the hard middle ground — and where FIM One stands.
- [Planning Landscape](https://docs.fim.ai/architecture/planning-landscape.md): Five kinds of planning in the AI tooling landscape and where FIM One fits.
- [Progressive Disclosure](https://docs.fim.ai/architecture/progressive-disclosure.md): How FIM One minimizes context consumption by loading tool definitions on demand -- stub-first, detail-on-request.
- [ReAct Engine](https://docs.fim.ai/architecture/react-engine.md): How FIM One's ReAct engine works — dual-mode execution, structured LLM calls, and streaming answer synthesis.
- [System Overview](https://docs.fim.ai/architecture/system-overview.md): How Agent, Knowledge Base, Connector, Built-in Tools, MCP, and Skills compose into a unified architecture.
- [Changelog](https://docs.fim.ai/changelog.md): All notable changes to FIM One.
- [AI Builder](https://docs.fim.ai/concepts/ai-builder.md): Use AI to build Connectors and Agents — quick suggestions or a full ReAct builder.
- [Artifacts](https://docs.fim.ai/concepts/artifacts.md): Browse, preview, and download all AI-generated files across your conversations.
- [Intelligent Document Processing](https://docs.fim.ai/concepts/document-processing.md): How FIM One processes uploaded documents for AI understanding, with automatic vision support for PDFs, Office files, and images.
- [Evaluation Center](https://docs.fim.ai/concepts/eval-center.md): Quantitatively benchmark your agents with test datasets and automated LLM grading.
- [Execution Modes](https://docs.fim.ai/concepts/execution-modes.md): Standalone, Copilot, and Hub — three ways to deploy FIM One.
- [Market](https://docs.fim.ai/concepts/market.md): Discover, subscribe to, and publish Solutions and Components across your organization and the global community.
- [Feishu (Lark) Channel](https://docs.fim.ai/configuration/channels/feishu.md): Wire FIM One to a Feishu app for outbound messages, inbound replies, and interactive approval cards.
- [Channels Overview](https://docs.fim.ai/configuration/channels/overview.md): Connect FIM One to external messaging platforms for notifications, approvals, and two-way agent interaction.
- [Database](https://docs.fim.ai/configuration/database.md): Choose and configure the right database backend for your deployment
- [Environment Variables](https://docs.fim.ai/configuration/environment-variables.md): Complete configuration reference for FIM One.
- [Model Management](https://docs.fim.ai/configuration/model-management.md): Configure LLM providers, models, and model groups in the admin panel.
- [Recommended Models](https://docs.fim.ai/configuration/models.md): Choose the right LLM for your FIM One deployment.
- [OAuth Providers](https://docs.fim.ai/configuration/oauth-providers.md): Configure GitHub, Google, Discord, and Feishu social login.
- [Contributing](https://docs.fim.ai/contributing.md): Development conventions, type safety, testing, and code quality standards for FIM One contributors.
- [Extension Guide](https://docs.fim.ai/extension-guide.md): How to plug custom LLM providers, tools, memory backends, embeddings, rerankers, web backends, and RAG components into FIM One.
- [FAQ](https://docs.fim.ai/faq.md): Frequently asked questions about FIM One — setup, features, licensing, and more.
- [Introduction](https://docs.fim.ai/index.md): AI-Powered Connector Hub — plug in systems, let AI orchestrate.
- [Exa](https://docs.fim.ai/integrations/exa.md): Use Exa as the default web search provider in FIM One agents — neural, fast, deep-reasoning, and instant modes with sensible defaults.
- [License](https://docs.fim.ai/license.md): FIM One Source Available License v1.2 — what you can and cannot do with the software.
- [Quick Start](https://docs.fim.ai/quickstart.md): Get FIM One running in minutes — or try the cloud version instantly.
- [Roadmap](https://docs.fim.ai/roadmap.md): Feature roadmap and version planning for FIM One.
- [Security](https://docs.fim.ai/security.md): How to report vulnerabilities, response timelines, and security best practices for self-hosted FIM One deployments.
- [Competitive Landscape](https://docs.fim.ai/strategy/competitive-landscape.md): Internal strategic reference — where FIM One sits among competitors.
- [Integration Paradigms](https://docs.fim.ai/strategy/integration-paradigms.md): CLI + Skills, MCP, or Connector Hub — understanding the three paradigms for connecting AI agents to enterprise systems.
- [Orthogonality Strategy](https://docs.fim.ai/strategy/orthogonality-strategy.md): Architecture Decision Record — classifying features by their relationship to LLM progress.
- [FIM One Whitepaper](https://docs.fim.ai/whitepaper.md): Data never leaves your boundary. The enterprise AI infrastructure layer, not another iPaaS.
- [Why FIM One](https://docs.fim.ai/why.md): Global enterprises run dozens of systems that don't talk to each other — across Slack, Teams, Feishu, WeCom, and everything in between. FIM One is the AI connector hub that bridges them all, without touching a single line of your code.

## OpenAPI Specs

- [openapi](https://docs.fim.ai/openapi.json)