Skip to content
Back to search

thclaws

Open-source AI agent harness in native Rust — GUI, CLI, headless, and webapp from one binary. Multi-provider, MCP, skills, plugins, agent teams.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/thclaws/thclaws with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/thclaws/thclaws'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/thclaws--thclaws/badge.svg)](/listings/thclaws--thclaws)

HTML

<a href="/listings/thclaws--thclaws"><img src="/api/listings/thclaws--thclaws/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

git clone https://github.com/thClaws/thClaws.git
cd thClaws

# One-shot: build frontend, then cargo build --features gui
./scripts/build.sh           # macOS / Linux
./scripts/build.ps1          # Windows PowerShell

# Verification suite (cargo fmt --check, clippy, tsc, cargo test)
./scripts/build.sh --check

# Run
cargo run --features gui              # GUI
cargo run -- --cli                    # CLI REPL
cargo run -- -p "explain crates/core" # one-shot

**Prerequisites:** current stable Rust (what CI builds on — nothing pins an MSRV), Node.js 20+, pnpm 10+ (`frontend/package.json` pins `packageManager: pnpm@10.x`, so corepack refuses 9). The helper enforces frontend-before-cargo order (the GUI build embeds `frontend/dist/index.html` at compile time via `include_str!`). See [CONTRIBUTING.md](CONTRIBUTING.md) for the full PR workflow.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 419 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <div align="center">
  2. # thClaws 🦞
  3. **The open-source agent harness, in your terminal and on your desktop.**
  4. A native-Rust AI agent workspace that codes, automates, remembers, and coordinates — running on your own machine. One binary. Sovereign by design. Built by a small team hacking in public.
  5. [![GitHub stars](https://img.shields.io/github/stars/thClaws/thClaws?style=social)](https://github.com/thClaws/thClaws/stargazers)
  6. [![Contributors](https://img.shields.io/github/contributors/thClaws/thClaws.svg)](https://github.com/thClaws/thClaws/graphs/contributors)
  7. [![Discussions](https://img.shields.io/github/discussions/thClaws/thClaws)](https://github.com/thClaws/thClaws/discussions)
  8. [![Release](https://img.shields.io/github/v/release/thClaws/thClaws)](https://github.com/thClaws/thClaws/releases)
  9. [![CI](https://img.shields.io/github/actions/workflow/status/thClaws/thClaws/release.yml?branch=main)](https://github.com/thClaws/thClaws/actions)
  10. [![License: MIT OR Apache-2.0](https://img.shields.io/badge/license-MIT%20OR%20Apache--2.0-blue.svg)](#license)
  11. [![Platform](https://img.shields.io/badge/platform-macOS%20·%20Windows%20·%20Linux-lightgrey.svg)](#installation)
  12. [![Built with Rust](https://img.shields.io/badge/built%20with-Rust-orange.svg)](https://www.rust-lang.org/)
  13. [Website](https://thclaws.ai) · [Download](https://thclaws.ai/downloads.html) · [Manual](https://thclaws.ai/manual) · [Enterprise](ENTERPRISE.md) · [Discussions](https://github.com/thClaws/thClaws/discussions) · [Contribute](#contribute) · [Supporters](SUPPORTERS.md)
  14. </div>
  15. ---
  16. ## 🆕 New in v0.126.0 — many agents, one workspace
  17. **A workspace is no longer one agent. It is a shelf of them.** Keep your coding agent, a research agent, and a book-writing agent side by side in the same project, and switch between them from a rail on the left without losing anything — each one keeps its own conversation, scroll position, and half-typed message while you are away.
  18. ```
  19. ┌────┬───────────────────────────────────────────────┐
  20. │ MA │ Chat · Terminal · Files · UI │
  21. │ RE │ │
  22. │ BA │ every agent: its own sessions, memory, │
  23. │ │ settings, and browser logins │
  24. │ + │ │
  25. └────┴───────────────────────────────────────────────┘
  26. ```
  27. - **Each agent is self-contained.** Its own sessions, knowledge bases, settings, and browser profile, in its own folder under `.thclaws/bots/<name>/`. An agent's file tools stop at that folder, so one agent cannot read or write another's work.
  28. - **They run side by side.** Every agent is its own process, supervised by the app. An agent that crashes is restarted without disturbing the others, and an agent you switch away from keeps working.
  29. - **Add an agent from the rail's `+`.** Get any Agent Template from [thClaws.cloud](https://thclaws.cloud/templates), or start a blank agent that behaves exactly like opening a new folder. Up to 8 agents run at once.
  30. - **Desktop and browser alike.** The same rail works in the desktop app and in `thclaws --serve`.
  31. **Your existing workspaces upgrade themselves.** The first time v0.126.0 opens a workspace — in the desktop app or with `--serve` — it moves your project into `.thclaws/bots/main/` (git history included) and tells you it did. Scripts keep working: `thclaws -p` and `--cli` at the workspace root step into that first agent automatically. An older thClaws opening an upgraded folder finds a note asking you to update, rather than a broken project.
  32. > **Before you upgrade:** anything that holds your project's old path — an editor window, a script with an absolute path, another clone — needs the new one. Commit or back up first if that matters to you. Your home directory is never upgraded, and hosted workspaces on thClaws.cloud are not migrated by this release.
  33. Full list of changes: [CHANGELOG.md](CHANGELOG.md)
  34. ---
  35. ## See it work
  36. Three tabs, one binary — captured from a live thClaws session looking at its own source.
  37. <div align="center">
  38. <a href="docs/img/screens-carousel.webp"><img src="docs/img/screens-carousel.webp" alt="thClaws Desktop GUI — cycling through Files, Terminal, and Chat tabs" width="900" /></a>
  39. **Files** (codemirror + tiptap) · **Terminal** (REPL · slash commands · ANSI tool output) · **Chat** (markdown render · tool indicators)
  40. </div>
  41. ---
  42. ## Highlights
  43. ### 🎨 Media Studio — generate images & video, built in
  44. A point-and-click shell for the new media tools — **Text → Image**, **Image Edit**, **Text → Video**, and **Image → Video** — across Google Gemini, OpenAI `gpt-image-2`, Alibaba Qwen, Veo, and DashScope HappyHorse. Pick a provider/model, set the resolution, hit Generate; finished media lands in a live gallery you can click to reuse as the source for the next run. The same tools (`TextToImage` / `ImageToImage` / `TextToVideo` / `ImageToVideo` / `MediaJobStatus`) are callable by the agent from chat too — async video included.
  45. <div align="center">
  46. <a href="docs/img/media-studio.webp"><img src="docs/img/media-studio.webp" alt="thClaws Media Studio — text-to-image with a live gallery of generated media" width="860" /></a>
  47. </div>
  48. ### 🧠 OpenRouter Fusion — many models, one answer
  49. `openrouter/fusion` fans your question out to a **panel** of up to 8 models in parallel (each can search the web), then a **judge** weighs the answers and synthesizes the consensus — multi-model deliberation packaged as a single model id. The new **`openrouter/fusion+`** makes it fully configurable right from the model picker: choose the panel, the judge, the per-model tool-call budget, temperature, reasoning effort, and tool-choice. A practical stand-in when the model you'd reach for isn't available on the platform.
  50. <div align="center">
  51. <a href="docs/img/fusion-config.webp"><img src="docs/img/fusion-config.webp" alt="thClaws — configuring the OpenRouter Fusion panel, judge, and limits" width="860" /></a>
  52. </div>
  53. Docs: [media tools](https://thclaws.ai/manual/ch11-built-in-tools.html) · [Media Studio](https://thclaws.ai/manual/ch26-gui-shells.html) · [Fusion setup](https://thclaws.ai/manual/ch06-providers-models-api-keys.html)
  54. ### Recently shipped
  55. A release most weeks. Full history in the [changelog](CHANGELOG.md); the
  56. things most likely to matter to you:
  57. - **Enterprise policy + client-side audit trail** — a signed org policy
  58. file turns the same binary into a managed client: branding, an
  59. extension allow-list, a forced LLM gateway, OIDC SSO, a forced
  60. permission mode, and a payload-free record of every tool call.
  61. ([ENTERPRISE.md](ENTERPRISE.md) · [ภาษาไทย](ENTERPRISE-th.md))
  62. - **OS-level Bash confinement, on by default** — shell commands run
  63. inside a kernel-enforced filesystem boundary (macOS Seatbelt, Linux
  64. Landlock, bubblewrap fallback), so an obfuscated command still can't
  65. write outside the workspace.
  66. - **Thai PII masking** — an opt-in pipeline that detects and tokenizes
  67. Thai personal data before it reaches the model, then restores it in
  68. the output.
  69. - **Movie Maker** — screenplay → shots → generated video, end to end.
  70. - **Browser automation** — drive a real Chrome session as a tool.
  71. - **Workflows** — deterministic multi-agent scripts (`/workflow run`).
  72. - **GUI Shells** — ship a custom HTML surface with your agent, themed
  73. and bridged to sessions, memory and tools.
  74. - **thclaws.cloud** — an agent catalog, a hosted runtime, and a metered
  75. gateway for people who don't want to hold provider keys.
  76. ---
  77. ## Hacking in public
  78. thClaws started in April 2026. Since then it has shipped **100+ public releases**, drawn **30+ contributors**, and lands roughly a release a week — the current release is **v0.121.0**. It's developed by a small team at **ThaiGPT Co., Ltd.** — and a meaningful chunk of the codebase comes from outside contributors who heard about it and stayed.
  79. We're aiming for **v1.0 = "the multi-platform agent"**: the same agent loop on your desktop, in your terminal, and bridged into Telegram, Discord, Slack, WhatsApp, Facebook Messenger, and LINE. Telegram, LINE, and Messenger are already shipping. Discord, Slack, and WhatsApp are next — and they're great places to plug in. ([Contribute →](#contribute))
  80. > Built in Thailand. Meant for the world.
  81. ---
  82. ## Four surfaces, one engine
  83. The same `Agent` loop, `Session`, and `ToolRegistry` back every UX:
  84. - **Desktop GUI** (`thclaws`) — native window with Terminal, Chat, Files, and optional Team tabs.
  85. - **CLI REPL** (`thclaws --cli`) — interactive terminal prompt for SSH, headless servers, or zero-GUI workflows.
  86. - **Non-interactive mode** (`thclaws -p "prompt"`) — single turn, exits. Pipe-friendly for scripts and CI. `-v` for token usage on stderr.
  87. - **Webapp** (`thclaws --serve --port 7878`) — same engine over WebSocket/HTTP, plus a drop-in API surface so other tools can drive the agent: **OpenAI-compatible** at `/v1/chat/completions` and **Anthropic-compatible** at `/v1/messages`. Point the openai or anthropic SDK at it, or set `ANTHROPIC_BASE_URL`. SSH-tunnel for "Claude Code anywhere" without opening a port.
  88. ---
  89. ## Features
  90. Everything's in one binary. Pick the surface that fits the task, swap the provider, drop in a skill, glue in an MCP server, then walk away while a scheduled job or background agent finishes the work.
  91. - **Multi-provider** — Anthropic (native + Claude Agent SDK via Claude Code auth), OpenAI (Chat Completions + Responses/Codex), Google Gemini & Gemma, Alibaba DashScope (Qwen), DeepSeek, Z.ai (GLM Coding Plan), NVIDIA NIM, NSTDA Thai LLM (OpenThaiGPT, Typhoon, Pathumma, THaLLE), xAI (Grok), Moonshot (Kimi), MiniMax, OpenRouter, Agentic Press, Azure AI Foundry, Ollama (local + Anthropic-compatible + Cloud), LMStudio, plus a generic **OpenAI-compatible** slot (`oai/*`) and first-class LiteLLM for internal proxies. The shipped catalogue carries **1,100+ models across 27 providers** — LLM plus image, video and audio. Switch mid-session with `/model` or `/provider`.
  92. - **Open standards, not a walled garden** — [Model Context Protocol](https://modelcontextprotocol.io/) for tools, [`AGENTS.md`](https://agents.md) for project instructions (adopted by Google, OpenAI, Factory, Sourcegraph, Cursor), `SKILL.md` with YAML frontmatter for packaged workflows. Configuration portable between thClaws, other compliant agents, and whatever comes next.
  93. - **Skills, plugins, MCP servers, hooks** — extend the agent without touching Rust. Skills are folders with a `SKILL.md`. Plugins bundle skills + commands + agent definitions + MCP servers under one manifest. MCP brings in third-party tools (GitHub, filesystems, browsers, Slack…) over stdio or HTTP-Streamable with OAuth 2.1+PKCE. Hooks run shell scripts on lifecycle events (`pre_tool_use`, `permission_denied`, `session_start`, …).
  94. - **Three tiers of agent orchestration** — model-driven subagents (`Task` tool, blocking, up to 3 levels deep); user-driven concurrent side-channels (`/agent <name> <prompt>`, parallel to main, own cancel token); multi-process **Agent Teams** with shared mailbox, task queue, tmux panes, and optional git worktrees.
  95. - **Knowledge bases (KMS) + `/dream`** — per-project and per-user wikis under `.thclaws/state/kms/<name>/pages/`, indexed by a one-line `index.md`. Grep + read (no embeddings), following Andrej Karpathy's LLM-wiki pattern. `/dream` mines your recent sessions in the background and writes a dated audit-trail page to review with `git diff`.
  96. - **Plan mode** — `EnterPlanMode` proposes an ordered list of steps you Approve / Cancel / Skip / Retry. Same UX in GUI sidebar and `/plan` slash command.
  97. - **Schedule recurring jobs** — `/schedule add` runs an agent on cron, fixed intervals, or filesystem changes (`watchWorkspace`). In-process scheduler for ephemeral, native daemon (`launchd` / `systemd-user`) for survives-reboot.
  98. - **Long-running loops & overnight builds** — `/loop` for fixed-interval iteration, `/goal` for audit-driven completion. `/goal --auto` is a Ralph-style overnight builder that keeps going until the goal is satisfied or you wake up.
  99. - **Document workflow** — native PDF, DOCX, PPTX, XLSX read + edit + create tools, plus image rendering. Ingest a 50-page PDF, summarize into KMS, produce a follow-up deck — one conversation.
  100. - **Memory & project instructions** — `AGENTS.md` (or `CLAUDE.md`) walked up from `cwd` and injected into the system prompt. Persistent memory store classified as `user` / `feedback` / `project` / `reference`, stored as markdown you can read, edit, or commit.
  101. - **Settings as one file** — `.thclaws/settings.json` (project) or `~/.config/thclaws/settings.json` (user). API keys go in the OS keychain by default (macOS Keychain / Windows Credential Manager / Linux Secret Service) with `.env` fallback for CI.
  102. - **Session resume** — `thclaws --resume last` or `--resume <id>`. Sessions live as JSONL under `.thclaws/state/sessions/` — git-friendly, grep-friendly, never opaque.
  103. - **Safety first** — shell commands run inside an **OS-enforced** filesystem boundary (macOS Seatbelt, Linux Landlock, bubblewrap fallback), so obfuscation doesn't get a command out of the workspace; modes `workspace` (default) / `strict` / `off`. On top of that: a filesystem sandbox scoped to the working directory, destructive commands flagged, and your approval on every mutating tool call unless you've opted into auto-approve. Permission requests label which agent is asking when multiple are running.
  104. - **Offline-capable** — Ollama (native + Anthropic-compatible) lets you run entirely against a local model. No cloud round-trip, no API key.
  105. - **Media, end to end** — Media Studio generates images and video (Gemini, OpenAI `gpt-image`, Qwen, Veo, LTX, iApp for Thai text) from a point-and-click shell or from chat; **Movie Maker** takes a screenplay through shots to a finished film.
  106. - **Browser automation** — drive a real Chrome session as a tool: navigate, read the page, fill forms, capture screenshots.
  107. - **Workflows & GUI Shells** — `/workflow run` scripts deterministic multi-agent pipelines; GUI Shells let you ship a custom HTML surface bridged to sessions, memory and tools, themed with the app.
  108. - **Thai PII masking (opt-in)** — detects and tokenizes Thai personal data before the model ever sees it, then restores it on the way out. Per-workspace toggle.
  109. - **Enterprise Edition — same binary, your rules** — a signed org policy file adds branding, an install allow-list, a forced LLM gateway, OIDC single sign-on, a forced permission mode and tool deny-list, and a payload-free audit trail of every tool call. No licence server, nothing phones home, and every control is in this repo. See [ENTERPRISE.md](ENTERPRISE.md) ([ภาษาไทย](ENTERPRISE-th.md)).
  110. - **thclaws.cloud** — an agent catalog, a hosted runtime, and a metered gateway, for people who'd rather not hold provider keys. Optional; the desktop binary never needs it.
  111. - **Deploy what you build** — ship landing pages, web apps, APIs, and AI agents through [Agentic Press Hosting](https://agentic-press.com) (partnered with SIS Cloud Service and Artech.Cloud) — or any host you prefer. Deploy flow ships as a plugin (`/plugin install …-deploy`), so hosts are swappable. The client never locks you in.
  112. - **Shell escape** — prefix any REPL line with `!` to run a shell command directly. No tokens, no approval prompt, no agent round-trip (`! git status`, `! ls`).
  113. ---
  114. ## Contribute
  115. **We'd love your help.** thClaws is built in the open by a small team and 30+ contributors so far. Reviews are typically fast, the codebase is approachable, and there's plenty of room to make a real dent.
  116. ### Quick start for contributors
  117. ```sh
  118. git clone https://github.com/thClaws/thClaws.git
  119. cd thClaws
  120. # One-shot: build frontend, then cargo build --features gui
  121. ./scripts/build.sh # macOS / Linux
  122. ./scripts/build.ps1 # Windows PowerShell
  123. # Verification suite (cargo fmt --check, clippy, tsc, cargo test)
  124. ./scripts/build.sh --check
  125. # Run
  126. cargo run --features gui # GUI
  127. cargo run -- --cli # CLI REPL
  128. cargo run -- -p "explain crates/core" # one-shot
  129. ```
  130. **Prerequisites:** current stable Rust (what CI builds on — nothing pins an MSRV), Node.js 20+, pnpm 10+ (`frontend/package.json` pins `packageManager: pnpm@10.x`, so corepack refuses 9). The helper enforces frontend-before-cargo order (the GUI build embeds `frontend/dist/index.html` at compile time via `include_str!`). See [CONTRIBUTING.md](CONTRIBUTING.md) for the full PR workflow.
  131. ### Where we need help right now
  132. If you want to land something impactful, these are the places we'd most love a hand:
  133. - 🚀 **Discord adapter** — bridge the agent into a Discord guild (same shape as our shipping Telegram + Messenger adapters in `crates/core/src/messenger/` and friends).
  134. - 🚀 **Slack adapter** — same idea, Slack-side.
  135. - 🚀 **WhatsApp adapter** — round out the v1.0 multi-platform thesis.
  136. - 🪟 **Windows ARM polish** — installer, GUI smoke-test on Surface / Snapdragon X.
  137. - 🧠 **KMS embeddings (opt-in)** — current KMS is grep + read; an optional embeddings layer alongside (not replacing) would be welcome.
  138. - 🌏 **i18n & translations** — manual chapters and UI strings are EN/TH today; we'd love help with more locales.
  139. - 🧩 **Skills, plugins, MCP servers** — no Rust needed. Build one, ship it on GitHub, list it on the marketplace. (See [Skills documentation](https://thclaws.ai/manual).)
  140. - 📝 **Docs, examples, walkthroughs** — typo fixes welcome, case studies even more welcome.
  141. Browse [**good first issues**](https://github.com/thClaws/thClaws/issues?q=is%3Aopen+is%3Aissue+label%3A%22good+first+issue%22) or [**help wanted**](https://github.com/thClaws/thClaws/issues?q=is%3Aopen+is%3Aissue+label%3A%22help+wanted%22) to start. If you want to land something bigger, drop into [Discussions](https://github.com/thClaws/thClaws/discussions) and pitch it first — that's the lowest-friction path.
  142. ### Codebase tour
  143. Everything ships from a single crate, `crates/core/`. The interesting modules:
  144. | Path | What it does |
  145. |---|---|
  146. | `crates/core/src/agent.rs` + `agent_runtime.rs` | The agent loop — turns, tool calls, streaming, cancellation |
  147. | `crates/core/src/commands.rs` | Slash-command registry (`/help`, `/model`, `/skill`, `/schedule`, …) |
  148. | `crates/core/src/kms.rs` | Knowledge bases (per-project + per-user) with `KmsRead`/`Search`/`Write`/`Append` tools |
  149. | `crates/core/src/memory.rs` | Persistent memory store (user / feedback / project / reference) |
  150. | `crates/core/src/mcp.rs` | MCP transport (stdio + HTTP-Streamable + OAuth 2.1 PKCE) |
  151. | `crates/core/src/hooks.rs` | Lifecycle hooks (`pre_tool_use`, `post_tool_use`, …) |
  152. | `crates/core/src/marketplace.rs` | Skill / plugin marketplace client |
  153. | `crates/core/src/permissions.rs` | Filesystem sandbox + tool approval flow |
  154. | `crates/core/src/compaction.rs` | Context-window management |
  155. | `crates/core/src/messenger/`, `line/` | Multi-platform adapters (the v1.0 thesis) |
  156. | `crates/core/src/confine.rs` | OS-level Bash confinement (Seatbelt / Landlock / bubblewrap) |
  157. | `crates/core/src/policy/` + `audit/` | Enterprise org policy (Ed25519 verify) and the tool-call audit trail |
  158. | `crates/core/src/media/` | Image / video generation providers behind the media tools |
  159. | `crates/core/src/api_v1/` | OpenAI-compatible HTTP API served by `--serve` |
  160. | `crates/core/src/gui.rs` + `ipc.rs` | wry/tao ↔ React bridge |
  161. | `frontend/src/` | React + Vite GUI (bundled into a single HTML file via `include_str!`) |
  162. Deeper engineering reference: [`thclaws-technical-manual/`](thclaws-technical-manual/) — agent loop, provider abstraction, KMS internals, side-channel + `/dream` plumbing, schedule daemon, hooks lifecycle, plan-mode driver, and the rest. **Read this before sending non-trivial PRs.**
  163. ### Contributors
  164. A huge thank you to the people who've shaped thClaws so far:
  165. <a href="https://github.com/thClaws/thClaws/graphs/contributors">
  166. <img src="https://contrib.rocks/image?repo=thClaws/thClaws" alt="thClaws contributors" />
  167. </a>
  168. ---
  169. ## Installation
  170. ### Build from source
  171. The fastest path if you want to contribute or just see how it's built. See [Contribute → Quick start](#quick-start-for-contributors).
  172. ### Pre-built binaries
  173. Download for your platform from the [Releases page](https://github.com/thClaws/thClaws/releases) or [thclaws.ai/downloads](https://thclaws.ai/downloads.html).
  174. Supported: macOS (Apple Silicon & Intel), Windows (x86_64 & ARM64), Linux (x86_64 & ARM64).
  175. <details>
  176. <summary><b>Linux runtime dependencies</b> — needed for the GUI binary on headless servers</summary>
  177. The Linux GUI binary links against Wayland and webkit2gtk at runtime. Desktop distros (Ubuntu Desktop, Fedora Workstation) ship these by default. **Headless servers** (cloud VMs, EC2, Docker images without a display) typically don't — `thclaws` will fail at startup with `error while loading shared libraries: libwayland-client.so.0`.
  178. Two options:
  179. **(a) Use CLI mode** — no GUI deps required:
  180. ```sh
  181. thclaws --cli
  182. thclaws -p "what does src/main.rs do?"
  183. ```
  184. **(b) Install the GUI deps**:
  185. ```sh
  186. # Debian / Ubuntu
  187. sudo apt install libwayland-client0 libwebkit2gtk-4.1-0 libsoup-3.0-0
  188. # Fedora / RHEL
  189. sudo dnf install wayland libsoup3 webkit2gtk4.1
  190. ```
  191. </details>
  192. ---
  193. ## Quick start
  194. ```sh
  195. # First run: pick a secrets backend (OS keychain or .env) when prompted
  196. thclaws
  197. # Configure a provider (inside the REPL)
  198. ❯ /provider anthropic
  199. ❯ /model claude-sonnet-5
  200. # Or try OpenRouter for 350+ models via one key
  201. ❯ /provider openrouter
  202. ❯ /model openrouter/anthropic/claude-sonnet-5
  203. # Drop an AGENTS.md or CLAUDE.md in your repo — it's read automatically
  204. # Useful slash commands
  205. ❯ /help # list everything
  206. ❯ /models # list available models for the current provider
  207. ❯ /kms # list attached knowledge bases
  208. ❯ /skill install https://github.com/anthropics/skills.git
  209. ❯ /mcp add github https://mcp.github.com
  210. ❯ ! git status # shell escape
  211. # Concurrent and long-running work
  212. ❯ /agent translator แปลไฟล์ src/foo.md เป็นภาษาไทย # spawn a side-channel agent
  213. ❯ /agents # list active background agents
  214. ❯ /dream # consolidate KMS in the background
  215. ❯ /schedule add --cron "0 9 * * MON-FRI" "review the day's PRs"
  216. # Headless mode
  217. thclaws -p "summarize CHANGELOG.md" # one-shot to stdout
  218. thclaws -p "summarize CHANGELOG.md" -v # + token usage on stderr
  219. thclaws --resume last # pick up the latest session
  220. # Web access
  221. thclaws --serve --port 7878 # then ssh -L 7878:localhost:7878 user@remote
  222. ```
  223. ---
  224. ## Configuration
  225. Settings are read in this precedence (higher wins):
  226. 1. CLI flags
  227. 2. `.thclaws/settings.json` (project)
  228. 3. `~/.config/thclaws/settings.json` (user)
  229. 4. `~/.claude/settings.json` (fallback)
  230. 5. Compiled-in defaults
  231. Open-standard files honored directly:
  232. - `CLAUDE.md` / `AGENTS.md` — system prompt additions, walked up from `cwd`
  233. - `.thclaws/skills/` / `.claude/skills/` — skill catalog
  234. - `.thclaws/agents/` / `.claude/agents/` — subagent definitions
  235. - `.mcp.json` / `.thclaws/mcp.json` — MCP server configuration
  236. - `.thclaws-plugin/plugin.json` / `.claude-plugin/plugin.json` — plugin manifest
  237. API keys are **never stored in config files** — only the OS keychain (default) or `.env`.
  238. ---
  239. ## Enterprise
  240. The same open-source binary becomes a managed client when it finds a
  241. **signed organization policy file** — no separate build, no licence
  242. check, nothing phoning home. The policy is signed with your own Ed25519
  243. key, verified against a public key compiled into your build, and a
  244. policy that has been edited stops the program rather than silently
  245. degrading.
  246. What a policy can set:
  247. | Block | What it does |
  248. |---|---|
  249. | `branding` | Your product name, logo and support contact, including in the agent's own system prompt |
  250. | `plugins` | Which hosts plugins, skills and HTTP MCP servers may be installed from |
  251. | `gateway` | Force every cloud model call through your own gateway (LiteLLM, Portkey, Azure OpenAI, in-house). Users' personal API keys are ignored |
  252. | `sso` | OIDC single sign-on against your IdP, so joiner/leaver applies to AI usage too |
  253. | `audit` | A payload-free record of every tool call — who, what, approved how, files touched — to a file or your SIEM |
  254. | `runtime` | Force the permission mode, deny named tools, switch off remote access and `--serve` |
  255. Every one of these controls is in this repository. The commercial part
  256. is the packaging — a build compiled to trust your key, signing
  257. infrastructure, deployment templates and support.
  258. 📖 **[Administrator guide → ENTERPRISE.md](ENTERPRISE.md)** ·
  259. 🇹🇭 **[ฉบับภาษาไทย → ENTERPRISE-th.md](ENTERPRISE-th.md)**
  260. Commercial enquiries: [enterprise@thaigpt.com](mailto:enterprise@thaigpt.com)
  261. ---
  262. ## Community
  263. - 💬 **[GitHub Discussions](https://github.com/thClaws/thClaws/discussions)** — questions, ideas, show-and-tell. The best place to start.
  264. - 🐛 **[Issues](https://github.com/thClaws/thClaws/issues)** — bug reports + concrete feature requests.
  265. - ✉️ **Email** — for security disclosures or commercial inquiries: [jimmy@thaigpt.com](mailto:jimmy@thaigpt.com) (see also [SECURITY.md](SECURITY.md)).
  266. ---
  267. ## Supporters
  268. In 2026 a mini fundraiser — selling of the e-book *Building thClaws* — helped fund this project. A huge thank-you to the **155 founding supporters** who backed it early: see the [**Supporters wall →**](SUPPORTERS.md). 🦞🙏
  269. ---
  270. ## Documentation
  271. - **Official site** — [thclaws.ai](https://thclaws.ai)
  272. - **Full user manual** — [thclaws.ai/manual](https://thclaws.ai/manual) or [`user-manual/`](user-manual/) (EN) / [`user-manual-th/`](user-manual-th/) (ภาษาไทย) — 30 chapters plus a providers/models/prices appendix. Applied case-study walkthroughs are in progress and land chapter by chapter.
  273. - **Technical manual** — [`thclaws-technical-manual/`](thclaws-technical-manual/) — engineering reference.
  274. - [Contributing](CONTRIBUTING.md) — dev setup, PR flow, commit style
  275. - [Changelog](CHANGELOG.md) — version history
  276. - [Code of Conduct](CODE_OF_CONDUCT.md) — Contributor Covenant 2.1
  277. - [Security](SECURITY.md) — vulnerability disclosure
  278. - [Enterprise](ENTERPRISE.md) ([ภาษาไทย](ENTERPRISE-th.md)) — signed org policy, gateway enforcement, SSO, audit trail, deployment runbook
  279. For books, training, and commercial deployment, see [agentic-press.com](https://agentic-press.com).
  280. ---
  281. ## License
  282. Dual-licensed under either:
  283. - [MIT License](LICENSE-MIT)
  284. - [Apache License 2.0](LICENSE-APACHE)
  285. at your option. Contributions are accepted under the same dual license — you keep your copyright; we just need permission to ship it.
  286. ---
  287. ## About
  288. thClaws is developed by **ThaiGPT Co., Ltd.** and published under a dual MIT/Apache-2.0 license. The client is free and open source forever. Enterprise Edition, hosting, and support are commercial offerings — see [agentic-press.com](https://agentic-press.com) or contact [jimmy@thaigpt.com](mailto:jimmy@thaigpt.com).
  289. > Built in Thailand. Meant for the world.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.