One API key for 70+ AI models. Route to GPT, Claude, Gemini, Grok and more
ReviewAudited by ClawScan on May 1, 2026.
Overview
This appears to be a disclosed LLM-gateway client; the main things to weigh are trusting the AIsa endpoint, protecting the API key, and avoiding sensitive prompts or images unless you accept the provider’s data handling.
This skill is purpose-aligned for routing LLM calls through AIsa. Before installing or using it, make sure you trust the AIsa/OpenClaw service with your prompts, images, and billing authority; keep AISA_API_KEY private; and avoid sending confidential data unless the provider’s privacy, retention, and routing terms are acceptable.
Findings (3)
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.
Anyone or any agent with this key can make requests under the user’s gateway account, potentially consuming quota or incurring charges.
The client reads the AISA_API_KEY environment variable and sends it as a bearer token to the gateway API. This is expected for the skill’s purpose, but it grants access to the user’s AIsa account.
self.api_key = api_key or os.environ.get("AISA_API_KEY") ... "Authorization": f"Bearer {self.api_key}"Keep AISA_API_KEY secret, scope or rotate it if the provider supports that, and revoke it if it may have been exposed.
Sensitive prompts, documents pasted into prompts, or image URLs submitted through this skill may be visible to the gateway service and, depending on routing, downstream model providers.
The client sends the selected model name and conversation messages to an external gateway endpoint. The skill also documents routing across multiple model families, so prompts or image URLs may be processed outside the local agent environment.
BASE_URL = "https://api.aisa.one/v1" ... payload = {"model": model, "messages": messages, "stream": stream}Use this only with data you are comfortable sending to the AIsa gateway and its routed providers; avoid confidential or regulated data unless the provider’s terms and retention controls meet your needs.
Users have less independent provenance information for verifying who maintains the skill or whether the endpoint and documentation match an official provider release.
The registry metadata does not identify a source repository or provenance source. No malicious install behavior is shown, but provenance matters because the skill handles an API credential and sends requests to an external service.
Source: unknown
Verify the OpenClaw/AIsa provider identity and endpoint before relying on the skill for sensitive or billable workloads.