cli-reference
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes a 'Task' section that interpolates user-supplied input via the '$ARGUMENTS' variable into the agent's context.\n
- Ingestion points: User input is ingested at the end of the SKILL.md file through the $ARGUMENTS placeholder.\n
- Boundary markers: There are no delimiters (e.g., XML tags or triple quotes) or 'ignore' instructions surrounding the user input to prevent it from being interpreted as agent commands.\n
- Capability inventory: The skill allows access to high-privilege tools including 'Bash', 'Write', 'Edit', and 'Read'.\n
- Sanitization: No sanitization or validation of the input is performed before interpolation.\n- [DATA_EXFILTRATION]: The documentation explicitly maps sensitive file paths used by the CLI, such as '~/.config/evenhub/credentials.yaml'. While this is legitimate reference material for the vendor's tool, it provides an agent or potential attacker with the exact location of authentication tokens and session data.\n- [EXTERNAL_DOWNLOADS]: The skill documents the installation of the '@evenrealities/evenhub-cli' package from the npm registry. This is a vendor-owned resource used to facilitate the primary purpose of the skill.
Audit Metadata