← Back to Skills Marketplace
genortg

Openclaw Comfyui Api Runner

by Krzysztof · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
88
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install openclaw-comfyui-api-runner
Description
Safe ComfyUI image generation. Use saved or ad-hoc server profiles, paste/upload raw workflow JSON, submit, poll, and serve downloaded outputs locally.
Usage Guidance
This skill appears to be what it says: a local ComfyUI workflow submitter and downloader. Before installing, consider: 1) The viewer server binds to 0.0.0.0 (not just localhost), so if you start the viewer its generated/ folder may be reachable from other machines on your network — set COMFYUI_SERVE_PORT and firewall rules or modify code to bind to 127.0.0.1 if you want it strictly local. 2) Saved profiles (including optional API key, username, password) are stored in plaintext JSON at ~/.config/openclaw/comfyui-runner.json — avoid saving secrets there if that is a concern and prefer supplying COMFYUI_API_KEY via ephemeral env at runtime. 3) The skill executes bundled Python scripts (no external downloads), so review the included files (already provided) if you need assurance. 4) Only point the runner at trusted ComfyUI instances — it will send workflow JSON and download images from the configured base_url. If you want me to, I can point out the exact lines that implement the viewer binding and config persistence so you can change them.
Capability Analysis
Type: OpenClaw Skill Name: openclaw-comfyui-api-runner Version: 1.0.0 The skill bundle is a legitimate tool for interacting with ComfyUI, providing features for profile management, workflow execution, and result viewing. It stores server configurations and credentials in a local JSON file (~/.config/openclaw/comfyui-runner.json) and includes a self-daemonizing HTTP server (scripts/serve_generated.py) to browse generated images. While the server binds to all network interfaces and stores credentials in plaintext, these are functional design choices rather than malicious indicators. The code implements basic security measures like path traversal checks in the file server and avoids dangerous execution patterns or unauthorized data exfiltration.
Capability Tags
requires-sensitive-credentials
Capability Assessment
Purpose & Capability
Name/description match the implementation: the scripts submit ComfyUI workflows, poll results, download outputs, and optionally serve them. The declared primary credential (COMFYUI_API_KEY) is used as an Authorization header. Minor mismatch: SKILL.md says it 'serves downloaded outputs locally' but the HTTP viewer binds to 0.0.0.0, making the files reachable from the network/LAN rather than strictly localhost.
Instruction Scope
SKILL.md directs the agent to run the included comfy.sh entrypoint which runs the Python runner. The runtime instructions and code limit actions to reading/writing the skill folder, a small per-user config (~/.config/openclaw/comfyui-runner.json), and talking to the configured ComfyUI base_url. The instructions do not request unrelated system files or unrelated credentials.
Install Mechanism
There is no remote install step — all code is bundled and executed locally. No downloads from third-party URLs or archive extraction are used. The only runtime requirement is python3, as declared.
Credentials
Required/declared credential (COMFYUI_API_KEY) is appropriate for the stated purpose and is actually used. The skill also supports COMFYUI_USERNAME and COMFYUI_PASSWORD and will persist credentials in ~/.config/openclaw/comfyui-runner.json if you save a profile — storing secrets on disk (in clear JSON) may be undesirable for some users. No unrelated environment variables or external service keys are requested.
Persistence & Privilege
always:false and the skill does not attempt to modify other skills or global agent settings. It does persist user profiles and defaults to ~/.config/openclaw/comfyui-runner.json (normal for a CLI tool) and may spawn a short-lived HTTP server to serve generated/ files. The server auto-exits after a timeout but is started as a subprocess accessible while running.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install openclaw-comfyui-api-runner
  3. After installation, invoke the skill by name or use /openclaw-comfyui-api-runner
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of comfyui-safe-connector, a safe ComfyUI image generation runner. - Supports saving and managing multiple ComfyUI server profiles. - Accepts workflows as pasted JSON, uploaded files, stdin, or built-in templates. - Submits and polls workflow jobs, downloading all output images to a local directory. - Provides output file paths and browser view URLs in the result JSON. - Prioritizes safe handling by parsing JSON only and not interpolating shell input.
Metadata
Slug openclaw-comfyui-api-runner
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Openclaw Comfyui Api Runner?

Safe ComfyUI image generation. Use saved or ad-hoc server profiles, paste/upload raw workflow JSON, submit, poll, and serve downloaded outputs locally. It is an AI Agent Skill for Claude Code / OpenClaw, with 88 downloads so far.

How do I install Openclaw Comfyui Api Runner?

Run "/install openclaw-comfyui-api-runner" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Openclaw Comfyui Api Runner free?

Yes, Openclaw Comfyui Api Runner is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Openclaw Comfyui Api Runner support?

Openclaw Comfyui Api Runner is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Openclaw Comfyui Api Runner?

It is built and maintained by Krzysztof (@genortg); the current version is v1.0.0.

💬 Comments