The Agent Skills Directory

[SAFE]: No security issues detected. The skill follows standard operating procedures for talent acquisition and includes benign Python scripts for metric analysis and document generation.
[COMMAND_EXECUTION]: The skill provides Python scripts (candidate_pipeline_tracker.py, interview_scorecard.py, job_posting_analyzer.py) for local execution. Analysis of these scripts confirms they perform purely analytical tasks (CSV parsing, regex-based text analysis) without invoking shell commands or accessing external network resources.
[DATA_EXPOSURE]: Script file access is limited to reading user-provided input files (CSV and markdown). There is no access to sensitive system directories (~/.ssh, ~/.aws) or environment variables containing credentials.
[PROMPT_INJECTION]: The instructions in SKILL.md define a senior talent acquisition role and workflow. There are no patterns suggesting attempts to bypass safety filters, override system instructions, or extract internal configuration.

talent-acquisition