← Back to Skills Marketplace
tokenrouter-image-generator
by
PaleBlueDot AI
· GitHub ↗
· v1.0.2
· MIT-0
120
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install tokenrouter-images
Description
Generate or edit images through Palebluedot Ai(PBD)-TokenRouter's multimodal image generation endpoint (`/v1/chat/completions`) using TokenRouter-compatible...
Usage Guidance
This skill appears to implement TokenRouter image generation as described, but note three issues before installing: (1) the package metadata fails to declare that it requires a TokenRouter API key — the script will look for PBD_TOKENROUTER_API_KEY or a --api-key and will also try to read ~/.openclaw/openclaw.json to auto-detect a key; (2) avoid pasting your API key into the chat as suggested in the README — that exposes the secret in chat logs and to the agent; prefer setting PBD_TOKENROUTER_API_KEY in your shell or configuring it in a secure place; (3) the script will write images to ~/.openclaw/media/outbound and will send image bytes and your API key to https://api.tokenrouter.com — only proceed if you trust TokenRouter and are comfortable with network access. If you want higher assurance, ask the author to: (a) update the skill metadata to declare required env vars, (b) remove or warn clearly against pasting keys into chat, and (c) document exactly what fields are read from ~/.openclaw/openclaw.json. If you accept those caveats this skill is functionally coherent; otherwise treat it as potentially risky.
Capability Analysis
Type: OpenClaw Skill
Name: tokenrouter-images
Version: 1.0.2
The skill bundle provides a legitimate tool for generating and editing images using the TokenRouter API. The script `scripts/generate_image.py` correctly implements image-to-image and text-to-image workflows, including a documented mechanism to retrieve API keys from environment variables or the local OpenClaw configuration file (~/.openclaw/openclaw.json). No evidence of malicious intent, unauthorized data exfiltration, or prompt injection attacks was found; the behavior is transparently described in SKILL.md and aligned with the tool's stated purpose.
Capability Tags
Capability Assessment
Purpose & Capability
The code and SKILL.md implement TokenRouter image generation and editing and call https://api.tokenrouter.com/v1/chat/completions as described — this matches the stated purpose. However the skill metadata declares no required environment variables or primary credential even though both SKILL.md and the script require a TokenRouter API key (PBD_TOKENROUTER_API_KEY) or a --api-key. The omission in the declared requirements is an incoherence.
Instruction Scope
Runtime instructions and the script read ~/.openclaw/openclaw.json to auto-detect an API key, create and write files under ~/.openclaw/media/outbound, and send image data (base64) to the external TokenRouter endpoint. The SKILL.md also explicitly advises users they can paste their API key into the chat for the agent to pass to the script — this encourages disclosing a secret to the agent and may result in the key being logged/stored in chat history. While reading openclaw.json and using ~/.openclaw paths is coherent with an OpenClaw skill, the SKILL.md grants the agent broad discretion to obtain the key from multiple locations (arg/env/config) which increases exposure.
Install Mechanism
This is an instruction-only skill with a small helper script; there is no install spec and nothing is downloaded or installed. Risk from install mechanism is low.
Credentials
The script reasonably needs a TokenRouter API key to call the service, but the skill metadata does not declare this required environment variable. The script also reads ~/.openclaw/openclaw.json to find apiKey values, which could expose stored provider credentials if present. Additionally, SKILL.md recommends pasting the API key into chat — that practice is disproportionate from a secrecy standpoint because it encourages exposing a secret to the agent and chat logs rather than setting a local env var or secure secret storage.
Persistence & Privilege
always: false and there is no install-time modification of other skills or system-wide settings. The skill writes output images to the OpenClaw media folder (expected behavior) and runs as-invoked. No elevated persistence or cross-skill config changes are present.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install tokenrouter-images - After installation, invoke the skill by name or use
/tokenrouter-images - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
- Added a list of additional supported image models (OpenAI and Google) in the Model + Image Config section, including explicit model IDs.
- Marked `google/gemini-2.5-flash-image` as the default/recommended model.
- Clarified that users may request a model switch using the listed TokenRouter-supported models.
- No changes to script functionality or interface—documentation only.
v1.0.1
- Adds auto-detection of TokenRouter API key from current agent's provider config in `~/.openclaw/openclaw.json` as a fallback after command-line and environment variable.
- Updates documentation to reflect new API key lookup priority and preflight logic.
- Improves error guidance when no API key is found, referencing the additional possible source.
- No changes to usage or command-line interface; existing integrations remain compatible.
v1.0.0
tokenrouter-images 1.0.0
- Initial release introduces the TokenRouter image generator skill for OpenClaw.
- Generate or edit images using TokenRouter-compatible models via the multimodal `/v1/chat/completions` endpoint.
- Requires a TokenRouter API key (set via argument or environment variable).
- Provides clear usage instructions for both text-to-image and image-to-image tasks.
- Supports custom model selection and provider-specific image configuration options (aspect ratio, image size).
- Includes extensive guidance for troubleshooting, preflight checks, and handling missing API keys.
Metadata
Frequently Asked Questions
What is tokenrouter-image-generator?
Generate or edit images through Palebluedot Ai(PBD)-TokenRouter's multimodal image generation endpoint (`/v1/chat/completions`) using TokenRouter-compatible... It is an AI Agent Skill for Claude Code / OpenClaw, with 120 downloads so far.
How do I install tokenrouter-image-generator?
Run "/install tokenrouter-images" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is tokenrouter-image-generator free?
Yes, tokenrouter-image-generator is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does tokenrouter-image-generator support?
tokenrouter-image-generator is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created tokenrouter-image-generator?
It is built and maintained by PaleBlueDot AI (@yb98k999); the current version is v1.0.2.
More Skills