wallet-integration
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands, including
parafor project management,npmfor dependency installation, andjqfor JSON configuration updates, as seen inSKILL.mdandreferences/para-workflows.md. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to interact with and process output from the
paraCLI, which could be manipulated by a malicious project configuration to influence agent behavior. - Ingestion points: The agent is instructed to read and act on output from commands like
para doctor,para keys list, andpara keys config showinreferences/para-workflows.mdandreferences/para-cli.md. - Boundary markers: There are no specific instructions to use delimiters or ignore instructions when processing output from the CLI tools.
- Capability inventory: The skill has the capability to write to the file system (creating
.pararc, updating.env.local, modifyingtsconfig.json) and execute further shell commands. - Sanitization: No mechanisms for sanitizing or validating the content of the CLI output are specified before the agent uses that data in subsequent operations.
Audit Metadata