erc8004
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts external data such as metadata URIs, JSON metadata, and feedback comments which are stored on the Stacks blockchain.
- Ingestion points: The
register,give-feedback, andrequest-validationcommands inerc8004.tsaccept untrusted string and JSON inputs from the user or external sources. - Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious instructions embedded in the URI or comment fields.
- Capability inventory: The skill has the capability to perform wallet-authorized blockchain transactions and write data to a decentralized ledger as seen in
erc8004.ts. - Sanitization: While the skill performs basic format validation (e.g., JSON parsing, hex normalization in
normalizeHex), it does not sanitize natural language content for prompt injection patterns before storage.
Audit Metadata