zest-asset-deposit-primitive
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill broadcasts transactions to the Stacks blockchain to perform DeFi operations.
- Evidence: Implementation of
makeContractCallandbroadcastTransactioninzest-asset-deposit-primitive.ts. - [EXTERNAL_DOWNLOADS]: The skill communicates with the Hiro API to fetch blockchain data and contract interfaces.
- Evidence: Requests to
https://api.hiro.sofound inzest-asset-deposit-primitive.ts. - Context: Interacts with well-known infrastructure services for the Stacks ecosystem.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied arguments and external API data to construct blockchain transactions, creating a potential injection surface.
- Ingestion points: CLI arguments (
--wallet,--amount,--deposit-asset) and JSON responses fromapi.hiro.so. - Boundary markers: Strict input parsing (e.g.,
parsePositiveBigInt), mandatory manual confirmation (--confirm=DEPOSIT), and read-only verification of protocol state before execution. - Capability inventory: Transaction broadcasting capabilities.
- Sanitization: Input validation, asset resolution against a fixed configuration, and enforcement of Stacks Post-Conditions to restrict the scope of asset movement.
- [CREDENTIALS_UNSAFE]: The skill reads session data and signing keys from the local agent environment to authorize transactions.
- Evidence: Accesses files in
~/.aibtc/sessions/and~/.aibtc/config.jsoninzest-asset-deposit-primitive.ts. - Context: These paths represent the standard credential storage for the AIBTC agent framework.
Audit Metadata