stacking
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides a CLI interface with subcommands (
stack-stx,extend-stacking) that sign and broadcast blockchain transactions to lock user tokens. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data such as Bitcoin reward addresses and stacking amounts which are passed directly to command arguments.
- Ingestion points: Subcommand arguments in
stacking/stacking.tsfor operations requiring transactions. - Boundary markers: The skill relies on standard CLI argument parsing and does not include specific boundary markers or instructions to the agent to disregard malicious content embedded within the data used for parameters.
- Capability inventory: The tool can perform financial operations by locking STX tokens for up to 12 reward cycles (approximately 6 months).
- Sanitization: The implementation validates numeric inputs like lock periods and versions using
parseInt, but it does not perform explicit format validation (e.g., regex for hex strings) for thepox-address-hashbyteswithin the CLI wrapper component.
Audit Metadata