nft
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches SIP-016 metadata for NFTs which may contain arbitrary text strings provided by any NFT creator.
- Ingestion points: The
get-metadatacommand innft.tsretrieves data from external NFT contracts via the Stacks network and Hiro API. - Boundary markers: The instructions do not define delimiters or specific warnings to the agent to treat the metadata as untrusted data.
- Capability inventory: The skill includes a
transfercommand innft.tsthat allows the agent to perform write operations (transferring assets) on the blockchain. - Sanitization: There is no evidence of sanitization or filtering of the strings retrieved from the metadata before they are passed back to the agent in JSON format.
- [COMMAND_EXECUTION]: The skill exposes the ability to perform asset transfers on a live blockchain via the
transfersubcommand. This is a high-impact capability that requires the user to maintain an unlocked wallet session and should be used with caution.
Audit Metadata