Cloud-hosted Agent access
External hardware and software can reach a cloud-hosted Buda Agent through REST resources under `/api/v1`, rather than requiring an on-device model.
API Claws
API Claws is Buda’s REST API layer for adding managed, cloud-hosted AI Agents to hardware, apps, browser extensions, and SaaS products. It provides Space-scoped sessions, Drive-based knowledge, and optional embedded chat without requiring developers to operate the Agent infrastructure themselves.
Managed AI Agents for connected products
API Claws is Buda's developer-facing REST API layer for adding a managed, cloud-hosted AI Agent to third-party hardware and software. It is intended for products such as smartwatches, earbuds, IoT devices, chat applications, browser extensions, and SaaS tools that need conversational AI without running a large model locally.
The integration combines model access, a managed Agent runtime, Drive-based knowledge, session management, tenant isolation, and wake/sleep orchestration. Developers can use API keys and Space-scoped resources to connect users or device fleets to Agents, maintain isolated conversations, and optionally expose chat through a short-lived embedded URL. Usage is charged through the Space credit system rather than a separate API subscription.
API Claw capabilities
Session-based conversations
Chat Session APIs provide per-user conversation context, asynchronous Agent runs, status, and messages. Each end user can receive an independent session with isolated context.
Drive-based knowledge
The Drive API supports durable files, manuals, policies, state, and long-lived memory for an Agent's knowledge base.
Space isolation
Space resources provide a tenant, customer, team, or device-fleet boundary. API requests authenticate with an API key and resolve to a Space before reaching the Agent.
On-demand runtime
Agents can sleep when idle and wake when a message arrives, so the connected product does not need to keep Agent infrastructure running continuously.
Embedded chat access
A backend can create a short-lived embed URL for a browser, mobile app, mini program, or extension. The returned token is kept in the URL hash and is used for frontend-safe `/api/v1/embed/...` calls.
Practical use cases
Add an assistant to smart hardware
A smartwatch, earbud, voice terminal, or connected IoT device can send input to a cloud Agent while the device handles its own input and output. Agent upgrades occur server-side instead of requiring the device to run a large model locally.
Extend a chat application
A chat application can attach an Agent to conversations and use independent Chat Sessions to keep each user's context separate.
Provide private AI support spaces
A customer-service SaaS can give each customer a private AI support space backed by customer-specific instructions or files, using Space boundaries and Drive-based knowledge.
Embed AI chat in a client interface
A browser extension, mobile app, mini program, or internal tool can use a backend-created short-lived embed URL so users can chat without exposing the main API key in the frontend.
Build a knowledge-backed product workflow
A product team can store manuals, policies, or other durable reference material in an Agent's Drive, allowing the Agent to use maintained knowledge instead of receiving a repeated large prompt.
Pros and Cons
Pros
- Provides a managed Agent layer instead of requiring developers to operate model infrastructure, runtime orchestration, retrieval, and session management separately.
- Supports separate context for end users through Chat Sessions and tenant or device-fleet boundaries through Spaces.
- Allows durable Agent knowledge to be maintained in a Drive rather than repeatedly included in prompts.
- Supports both backend-controlled API access and short-lived embedded chat access for frontend clients.
- Agents sleep while idle and wake when messages arrive, which suits products that need on-demand cloud AI.
Cons
- The supplied documentation does not include complete endpoint schemas, authentication examples for every operation, or detailed request and response behavior.
- API activity depends on the Space's available credit balance; AI features pause when the relevant credit pools are exhausted.
- A connected product still needs to build its own user experience and backend workflow around the API or embed surface.
FAQ
How does an application connect to API Claws?
API Claws lets a third-party product connect to a cloud-hosted Buda Agent through REST calls under `/api/v1`. The documented workflow includes authenticating with an API key, resolving a Space, reaching an Agent and its Drive, and running requests inside Chat Sessions. The page also documents an embed flow in which a backend creates a short-lived URL for a browser or app.
How are conversations and user context handled?
Each end user can have an independent Chat Session with isolated context. Sessions are associated with the Agent and support conversations, asynchronous Agent runs, status, and messages. The source does not provide a complete request and response schema for every endpoint.
Can an API Claw use a private knowledge base?
API Claws uses the Agent's Drive as a durable knowledge base for files such as manuals, policies, and other long-lived information. This avoids sending the same large prompt on every request, while file ingestion and knowledge-base processing are managed by Buda.
How is API Claw usage billed?
API Claw usage is billed through the Space credit system rather than a separate API subscription. Space credits are shared across Agents in that Space and are consumed according to model use and task complexity; when available credit pools are exhausted, AI features pause until credits reset or the Space is recharged.
Quick Facts
- Category
- Developer Tool; AI Chat
- API base
- REST resources under /api/v1
- Primary users
- Hardware makers, software developers, and SaaS providers
- Core workflow
- API key → Space → Agent and Drive → Chat Session
- Knowledge layer
- Agent Drive for files, manuals, policies, state, and long-lived memory
- Billing model
- Space credits; no separate API Claw subscription is described
Alternatives à API Claws
CreateOS Sandbox
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
GitBot
GitBot is a local browser-based workspace for creating reusable coding bots powered by Claude Code, Codex, or OpenCode. It helps developers run those agents across repository folders with persistent threads and configurable permissions.
AakarDev AI
AakarDev AI helps teams manage AI provider access, project-level setups, logs, and analytics from one dashboard. It supports BYOK workflows and lists providers including OpenAI, Google Gemini, Anthropic, Groq, Mistral AI, and Perplexity AI.
Trigger.dev chat agent
Trigger.dev chat agent is a durable AI chat backend for developers building stateful conversations that can survive refreshes, crashes, and long-running turns. It connects with the AI SDK `useChat` flow and runs on managed infrastructure with no timeout on a turn.
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
Codex Plugins
Codex Plugins bundle reusable skills, app integrations, and MCP servers into workflows you can install in the Codex app or use from Codex CLI. They help extend Codex with connected-service tasks, reusable instructions, and shared team workflows.