OpenAgent

A modern desktop AI agent client — built with Tauri, SvelteKit, and Rust.

English · 简体中文

Highlights

• Dual-Agent Architecture — A streaming Chat Agent for replies, and an async Memory Agent that distils long-term context into searchable records.
• Sub-Agent Delegation — The Chat Agent can call spawn_agent to delegate tasks to nested sub-agents; progress streams in real-time into a sub-conversation shown nested under the parent in the sidebar.
• Hybrid Long-Term Memory — SQLite + FTS5 + 384-dim embeddings (fastembed AllMiniLML6V2) blended with time decay for cross-session recall.
• Interactive User Prompts (ask_user) — The agent can pause mid-task and surface a structured form to the user — text, select, checkbox_group, confirm, date, and more. The agent blocks until the user responds, then continues with the collected values. No more one-shot guessing on ambiguous instructions.
• AGUI — Inline Interactive Components — The agent can embed interactive UI components directly in its prose: File(path: ...) chips that open files in-editor, Url(href: ...) capsules, Chart(type: bar|line|pie, ...) data visualisations, and Card(title: ...) summary panels — all rendered live by the streamdown engine.
• MCP-Native — Connect external MCP servers over HTTP or stdio; tools are injected into the agent at call time.
• First-class Dev Tools — Built-in read_file, write_file, edit_file, glob, grep, terminal_exec, web search (Brave / Tavily / SearXNG), and fetch via Jina Reader.
• Skills System — Drop a SKILL.md into ~/.config/openagent/skills/ or <workspace>/.agents/skills/ and the agent will discover it automatically.
• Checkpoints & File-Change Rollback — Every turn is a checkpoint with reverse diffs; undo a single file or rewind the whole agent.
• Pluggable LLMs — Anthropic Claude, OpenAI-compatible APIs, and anything you point a base URL at.
• Observability — Optional Langfuse tracing via OpenTelemetry (gen_ai.* attributes).
• Polished UI — Apple-style design language, streaming markdown, Mermaid & ECharts rendering, light/dark themes, i18n (zh / en).

Quick Start

Prerequisites

On Windows the Tauri prerequisites also include WebView2 and the MSVC build tools. See the official Tauri prerequisites for platform-specific setup.

Clone & install

git clone https://github.com/BANG404/openagent.git
cd openagent
bun install

Run in dev mode

# Full Tauri desktop app (frontend + Rust backend)
bun tauri dev

# OR — frontend only (port 14221, no Rust)
bun run dev

Build a distributable

bun tauri build

The built installers / app bundle land in src-tauri/target/release/bundle/.

Configure your first provider

On first launch OpenAgent creates ~/.config/openagent/config.toml. Open Settings → Providers and add a provider, or edit the file directly:

[[providers]]
id = "anthropic-main"
name = "Anthropic"
provider = "anthropic"
api_key = "sk-ant-..."
base_url = "https://api.anthropic.com"
enabled = true

[defaults]
chat_model   = { provider_id = "anthropic-main", model = "claude-sonnet-4-6" }
flash_model  = { provider_id = "anthropic-main", model = "claude-haiku-4-5" }

OpenAI-compatible endpoints (DeepSeek, OpenRouter, local Ollama, etc.) work the same — just point base_url at the right host and set provider = "openai".

Example: write a Skill

Create ~/.config/openagent/skills/python-review/SKILL.md:

---
name: python-review
description: Review Python diffs for type-hint coverage, error handling, and PEP 8 compliance.
---

When asked to review Python code:
1. Check that public functions have type hints.
2. Flag bare `except:` clauses and silent failures.
3. Suggest more idiomatic stdlib alternatives where appropriate.

That's it — OpenAgent picks it up on the next message and lists it in the agent's system prompt. The agent reads the full body on demand via read_file.

Interactive prompts with ask_user

When the agent needs a decision before continuing — ambiguous instruction, technology choice, destructive operation, missing parameter — it calls ask_user to surface a structured form in the chat panel. The agent blocks until you respond; everything in the form is typed, so you can select from dropdowns, tick checkboxes, confirm a boolean, or pick a date without typing a sentence.

Supported field types:

The agent is guided to ask once, ask clearly, and prefer structured fields over open text boxes.

AGUI — Inline Interactive Components

Beyond markdown, the agent can embed interactive components directly in its responses. The frontend's streamdown renderer picks them up and renders them as rich, clickable elements — no copy-pasting paths or URLs needed.

Syntax: ComponentName(prop: value, prop2: "string")

Multi-series charts use series: [{name, data}, ...]. The agent uses these by default whenever it references a project file, an external link, or presents comparative data — plain paths and bare URLs are treated as a fallback.

Memory file format

Memory files have two zones. The Memory Agent only writes below the marker comment:

## [User] Personal habits
<!-- You edit freely here; the agent never touches this section -->

## [Agent] Recent context summary
<!-- Memory Agent only operates below this comment -->

• Global memory → ~/.config/openagent/memory.md (every conversation)
• Local memory → <workspace>/.agents/memory.md (workspace-scoped)

Architecture at a glance

┌──────────────────────────────┐        Tauri IPC         ┌──────────────────────────────┐
│   SvelteKit Webview (src/)   │  ◄──────────────────►   │  Rust Core (src-tauri/)       │
│ ── routes/+page.svelte       │  invoke / event          │ ── commands.rs                │
│ ── lib/components/...        │  chat-chunk              │ ── tools.rs · terminal.rs     │
│ ── lib/chatStream.ts         │  chat-tool-call          │ ── conversation_memory.rs     │
│ ── lib/streamdown/...        │  chat-done               │ ── skills.rs · mcp.rs         │
└──────────────────────────────┘                          │ ── checkpoint.rs · state.rs   │
                                                          └──────────────────────────────┘
                                                                       │
                                          ┌────────────────────────────┼────────────────────────────┐
                                          ▼                            ▼                            ▼
                                ┌──────────────────┐         ┌──────────────────┐         ┌──────────────────┐
                                │  SQLite + FTS5   │         │   LLM Providers   │         │   MCP Servers     │
                                │  + fastembed     │         │  Claude · OpenAI  │         │  HTTP · stdio     │
                                └──────────────────┘         └──────────────────┘         └──────────────────┘

See CLAUDE.md for the full module-by-module breakdown.

Project structure

.
├── src/                      # SvelteKit frontend (Svelte 5 · TypeScript)
│   ├── routes/               # Page components
│   └── lib/                  # Components, stores, streamdown, types
├── src-tauri/                # Tauri / Rust backend
│   ├── src/
│   │   ├── lib.rs            # App setup, plugin registration
│   │   ├── commands.rs       # IPC command handlers
│   │   ├── tools.rs          # read_file, write_file, glob, grep, fetch (Jina), websearch, ...
│   │   ├── terminal.rs       # terminal_exec + background sessions
│   │   ├── sub_agent.rs      # spawn_agent tool — nested sub-agent execution
│   │   ├── conversation_memory.rs   # SQLite + fastembed hybrid retrieval
│   │   ├── checkpoint.rs     # Per-turn checkpoints + reverse diffs
│   │   ├── skills.rs         # SKILL.md discovery
│   │   ├── mcp.rs            # MCP transports + tool injection
│   │   └── context.rs        # System prompt assembly
│   └── icons/                # App icons (all platforms)
└── docs/                     # Design specs and architecture notes

Repository activity

Star history

Contributors

Thanks goes to these wonderful people:

Roadmap

• First-class screenshots & demo GIFs in this README
• Pre-built installers via GitHub Releases (alpha / beta / stable channels)
• In-app skill marketplace
• Multi-window / multi-workspace support

See open issues for the full backlog and feel free to file new ones.

Contributing

Contributions are very welcome — feature ideas, bug fixes, and docs improvements alike.

1. Fork the repo and create a branch from master.
2. Follow the Conventional Commits style — see existing log for the scopes we use (feat(toast):, fix(mermaid):, refactor(ui):, etc.).
3. Run bun run check and cargo check --manifest-path src-tauri/Cargo.toml before opening a PR.
4. Open a PR — describe why, not just what.

Project conventions live in CLAUDE.md; the UI/UX spec lives in docs/design.md.

Observability (optional)

Drop a .env in the project root to enable Langfuse tracing:

LANGFUSE_PUBLIC_KEY=pk-...
LANGFUSE_SECRET_KEY=sk-...
LANGFUSE_HOST=https://cloud.langfuse.com

When keys are present, Chat & Memory agent calls are instrumented with gen_ai.* OpenTelemetry attributes and exported via batch processor.

Further reading

• CLAUDE.md — Architecture reference for contributors and agents
• CHANGELOG.md — Full release history
• docs/design.md — Apple-style design spec
• Tauri docs · SvelteKit docs · rig (Rust LLM)

License

Released under the MIT License. © BANG404 and contributors.