dating-profile-doctor
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions and markdown. It does not include any scripts (.py, .js, .sh), binaries, or configuration files that could be used for code execution.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data from the user, including current dating profiles and personal text messages. Specifically, it requests 'paste a few texts to friends' and 'the current profile verbatim' in the Required Inputs section of SKILL.md. However, because the skill does not define or use any tools (allowed-tools is empty/missing), the risk is confined to the immediate conversation context with no path to system-level exploitation.
- [DATA_EXFILTRATION]: Although the skill asks for sensitive personal communication (text messages), there are no commands, network requests (curl, wget), or API call instructions that would allow this data to be sent to an external server.
- [COMMAND_EXECUTION]: There are no shell commands, backticks, or dynamic context injection patterns (e.g., Use of !
command) found in the file.
Audit Metadata