RomanNum3ral
/
openwebui-claude-code
mirror of https://github.com/tfriedel/openwebui-claude-code.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
Open WebUI Pipe: run Claude Code's agent loop inside chats via the Claude Agent SDK
7 Commits 1 Branch 0 Tags 115 KiB
Python 91.4%
Shell 5.6%
Dockerfile 3%
Go to file
Thomas Friedel c5f6764470 feat(sandbox): enable Claude Code built-in sandbox for Bash + WebFetch egress 
Closes #10. Until now the sandbox container had unrestricted outbound NAT
on sandbox-net — a compromised agent could curl arbitrary hosts, pip/npm
install malicious packages, or exfiltrate files via WebFetch. The
anthropic-proxy path-allowlist (#6) guarded only the credential-injection
proxy, not general egress.

Two layers added, both shipped as managed (enterprise) settings so user
and project settings cannot downgrade them:

- sandbox.*  → bubblewrap + local proxy gating Bash subprocesses (and all
  their children: pip, npm, git, curl). enableWeakerNestedSandbox=true
  because we run inside Docker without privileged user namespaces;
  accepted because the Docker container boundary is unchanged and this
  adds new egress filtering that previously did not exist.
- permissions.* → deny-by-default WebFetch with explicit domain allows
  (WebFetch bypasses the OS sandbox and needs its own allowlist).

failIfUnavailable=true + allowUnsandboxedCommands=false close the two
silent-fallback paths: if bubblewrap/socat install breaks, the container
fails loud instead of serving unsandboxed bash; and the
dangerouslyDisableSandbox escape hatch is disabled so sandbox violations
can't be bypassed via a permission prompt in our unattended deployment.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
 		2026-04-18 23:31:07 +02:00
anthropic-proxy feat(sandbox): route Anthropic traffic through credential-injecting proxy 2026-04-18 21:59:59 +02:00
sandbox feat(sandbox): enable Claude Code built-in sandbox for Bash + WebFetch egress 2026-04-18 23:31:07 +02:00
.gitignore Initial commit: Claude Code pipe for Open WebUI 2026-04-18 14:55:03 +02:00
README.md Disable fast path — always run full agent loop 2026-04-18 15:19:57 +02:00
claude_agent_pipe.py Disable fast path — always run full agent loop 2026-04-18 15:19:57 +02:00
claude_agent_pipe_sandboxed.py feat(sandbox): recover chat session mapping after OWUI restart 2026-04-18 23:04:25 +02:00
sync_pipe.py feat: sandboxed Claude Code pipe via open-terminal 2026-04-18 21:02:01 +02:00

README.md

OpenWebUI Claude Code Pipe

Run Claude Code's agent loop from inside Open WebUI chats, via the Claude Agent SDK.

This is an Open WebUI Pipe that exposes Claude Code as a selectable model. Each chat gets its own isolated workspace directory; agent turns within the same chat resume the same Claude Code session, so context (files, prior tool calls) carries forward.

Features

  • Full Claude Code agent loop — Read, Write, Edit, Bash, Glob, Grep, WebSearch, WebFetch (configurable allowlist)
  • Per-chat workspaces — each chat_id gets a sandboxed working directory that persists across turns
  • Dual auth — bring your own Anthropic API key (pay-per-token) or a Claude Pro/Max OAuth token (bills against your subscription)
  • Streaming UI — tool calls render inline with previews; generated images/PDFs/CSVs surface as artifacts in the chat
  • Configurable valves — model, permission mode, tool allowlist, max turns, workspace root

Requirements

  • Open WebUI (any recent version with the Pipes/Functions framework)
  • Python deps (auto-installed by Open WebUI from the file header):
    • claude-agent-sdk>=0.1.60
    • anthropic>=0.40.0
  • The claude CLI must be available on the host running Open WebUI's Python backend (the SDK shells out to it). Install via npm install -g @anthropic-ai/claude-code.

Installation

  1. In Open WebUI, go to Workspace → Functions → + (or Admin Panel → Functions).
  2. Paste the contents of claude_agent_pipe.py into the editor.
  3. Save and enable the function.
  4. Open the function's Valves and configure auth (one of):
    • ANTHROPIC_API_KEY — standard pay-per-token billing
    • CLAUDE_CODE_OAUTH_TOKEN — generate on a machine with a browser via claude setup-token; bills against your Pro/Max/Team subscription
  5. A new model named Claude Code will appear in the model picker.

Configuration (Valves)

Valve Default Description
ANTHROPIC_API_KEY (env) Anthropic API key. Falls back to the backend's env var.
CLAUDE_CODE_OAUTH_TOKEN (empty) Claude subscription OAuth token. Takes priority over the API key when set.
MODEL claude-haiku-4-5 Claude model ID (e.g. claude-haiku-4-5, claude-sonnet-4-6, claude-opus-4-7).
PERMISSION_MODE bypassPermissions default, acceptEdits, bypassPermissions, plan, or dontAsk.
ALLOWED_TOOLS Read,Write,Edit,Bash,Glob,Grep,WebSearch,WebFetch Comma-separated tools auto-approved without prompting.
WORKDIR_ROOT /tmp/claude-agent-pipe Root directory for per-chat workspaces.
MAX_TURNS 30 Max agent turns per user message. 0 disables the cap.

Auth notes

When both auth methods are present, the OAuth token wins and the API key is unset before invoking the SDK so it can't override.

Per Anthropic's terms: a Claude subscription is for personal use — don't re-offer subscription auth to other end users through a shared Open WebUI deployment. For multi-user setups, use API keys.

License

MIT