free quota text to image
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill matches its text-to-image purpose, but it sends prompts to external providers and can use and locally persist provider token state for quota rotation.
Install only if you are comfortable sending prompts to the listed third-party providers. Configure only the provider tokens you want this skill to use, keep the token-status file private, and leave the OpenAI-compatible paid fallback disabled unless you explicitly want that route.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Configured provider tokens may be used automatically for generation attempts and quota rotation, and token strings may appear in the local token-status file.
Provider API token values are used as persisted state keys when quota is exhausted, creating a local record tied to the configured credentials.
exhausted[token] = True
self._save_state()Configure only tokens you intend this skill to use, prefer limited-scope or disposable provider tokens, and protect or periodically clear the token-status state file.
Prompts, optimized prompts, and generation requests may be processed by external services such as Pollinations, Hugging Face, Gitee, ModelScope, A4F, or a configured OpenAI-compatible endpoint.
Prompt optimization sends the user's prompt text to an external text API before image generation, which is purpose-aligned but crosses a third-party data boundary.
{"role": "user", "content": prompt},
...
return _chat_request(POLLINATIONS_API_URL, payload, headers={}, timeout=timeout, fallback=prompt)Avoid confidential prompts unless you trust the selected providers; disable prompt optimization or force a specific provider when you need tighter data control.