The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the exec tool to perform initial setup by creating a local storage directory (mkdir -p ~/.openclaw/resume-builder). This is a legitimate functional requirement for storing user profiles and settings locally.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted job descriptions from users to perform features like resume tailoring and gap analysis.
Ingestion points: Job description text pasted by the user into the chat for Features 6 and 18.
Boundary markers: The skill does not define specific delimiters or instructional guardrails to prevent the AI from following commands potentially embedded within the job description text.
Capability inventory: The agent has access to read and write tools for the local storage directory, and a limited exec capability for directory management.
Sanitization: There is no evidence of text sanitization or filtering applied to the job descriptions before they are processed by the agent.

resume-builder