AISP

PassAudited by ClawScan on May 1, 2026.

Overview

The skill is coherent for renting or listing inference capacity, but it involves real USDC transactions and Venice API-key handling through an external backend, so users should verify details and approve actions deliberately.

Before installing or using this skill, verify the backend URL, SDK/project source, contract address, USDC address, rental terms, and all wallet prompts. Use a hardware or external wallet, never paste private keys, and only list revocable inference-only Venice API keys.

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.

What this means

A mistaken or poorly reviewed transaction could spend USDC, approve contract spending, or incur marketplace fees.

Why it was flagged

These instructions involve token approvals and on-chain funding. That is central to the skill's marketplace purpose, and the same file says to require explicit user confirmation, but it is still a high-impact action.

Skill content
Approve USDC if needed (contract spends on `fund`) ... Fund on-chain: `contract.fund(listingId, termDays, diemAmount)`
Recommendation

Confirm contract addresses, amounts, terms, and fees in the wallet before approving or funding any rental.

What this means

Using the wrong wallet or an over-privileged API key could expose funds or broader API account capabilities.

Why it was flagged

The workflow relies on wallet signing and provider API keys. These credentials are expected for the stated purpose, and the skill includes safety guidance, but they grant real account/API authority.

Skill content
POST /api/keys with `{ listingId, apiKey, signature, timestamp }` ... Use an external signer or hardware wallet; never paste raw private keys.
Recommendation

Use a hardware or external wallet, verify every signature message, never provide private keys, and only use scoped, revocable inference-only API keys.

What this means

A malicious or mistaken backend URL, or an over-scoped key, could expose API access beyond the intended rental use.

Why it was flagged

Sensitive API keys are transmitted to and from an external backend, and the balance endpoint is documented as accepting a raw API key in the request body. This is disclosed and purpose-aligned, but it creates a sensitive data boundary.

Skill content
`/api/keys` | POST | Provider stores API key ... `/api/key/:rentalId` | POST | Agent retrieves key ... `balance`: `apiKey` in body (no signature)
Recommendation

Use only a trusted HTTPS backend, verify BACKEND_URL, and send only scoped, revocable, inference-only keys.