create-profile

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill manages a local configuration file at ~/.claude/PROFILE.md. Access is strictly controlled via tool definitions (Read and Write) specifically scoped to this path, preventing unauthorized access to other parts of the filesystem.
  • [PROMPT_INJECTION]: The skill collects and stores user-provided information that is subsequently used as context for other skills. This creates a functional surface for indirect prompt injection, which is a necessary part of its role as a personalization tool.
  • Ingestion points: Information is collected through interactive questions using the AskUserQuestion tool.
  • Boundary markers: User input is stored in the PROFILE.md file using simple markdown headers without specific delimiters or isolation instructions.
  • Capability inventory: The skill has the ability to read and write to a specific configuration file in the user's home directory.
  • Sanitization: User-provided responses are written to the profile file as provided to maintain the user's intent and specific wording.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 03:03 PM
Security Audit — agent-trust-hub — create-profile