recruit-compare
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted external data (resumes, LinkedIn profiles, and interview notes) to generate a comparison report. An adversary could embed hidden instructions within a candidate's resume or profile to manipulate the agent's scoring, bias the final recommendation, or attempt to override the skill's defined rules.
- Ingestion points: Candidate information provided via the
<c1>and<c2>arguments, including resumes, LinkedIn profiles, and interview notes mentioned inSKILL.md. - Boundary markers: None. The skill does not instruct the agent to use delimiters or specific ignore-tags when processing candidate-provided content.
- Capability inventory: The skill generates structured text and writes output to a local file (
RECRUIT-COMPARE.md). - Sanitization: The skill lacks instructions for sanitizing, escaping, or validating the content of the ingested resumes or notes before interpolation into the comparison logic.
Audit Metadata