cli-anything
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands, such as tool help flags and verification runs, to generate accurate skill wrappers. This process relies on the agent's existing shell execution capabilities.- [INDIRECT_PROMPT_INJECTION]: The skill involves reading and parsing output from local command-line tools. Adversarial content within a tool's help text or output could potentially influence the agent's behavior during the introspection process.
- Ingestion points: Tool output (stdout/stderr) from local CLI commands is processed in Step 1 and Step 5 of the instructions.
- Boundary markers: There are no specific instructions to use delimiters or ignore directives when processing external tool output.
- Capability inventory: The skill utilizes the agent's ability to execute shell commands and write to the filesystem.
- Sanitization: The instructions do not specify a requirement to sanitize or validate the content of the tool output before processing.
Audit Metadata