applicant-profile-reader
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of Markdown documentation and a JSON evaluation file. It does not include any executable scripts, shell commands, or binary files.
- [DATA_EXPOSURE]: The skill is designed to process personal documents (CVs, transcripts, SOPs). However, it lacks any network capabilities (like curl or wget) or file-writing commands that would enable data exfiltration or unauthorized exposure. It functions as a text-processing assistant within the user's session.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user content from files like research notes or SOP drafts. This represents a potential surface for indirect prompt injection. The skill mitigates this by instructing the agent to 'Inventory the sources', 'Separate facts, evidence, and interpretation', and 'keep this synthesis grounded in the provided materials'. These instructions provide structural boundaries that help the model distinguish between instructions and data.
Audit Metadata