alchemy-core-workflow-b
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process NFT metadata and smart contract state from the blockchain, which are untrusted external sources.\n
- Ingestion points: Data is ingested through
alchemy.nft.getContractMetadata,alchemy.nft.getNftsForContract, andalchemy.nft.getNftMetadataBatchinsrc/nft/collection-explorer.tsandsrc/nft/batch-metadata.ts.\n - Boundary markers: The instructions do not define clear boundaries or delimiters for the agent when handling these external strings.\n
- Capability inventory: The skill possesses file modification capabilities (
Write,Edit) and package manager access (Bash(npm:*)), which could be targeted by a successful injection.\n - Sanitization: The logic directly maps external API responses to internal objects without filtering or sanitizing strings like descriptions or names.
Audit Metadata