The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized exfiltration were detected in the instructions or references.
[DATA_EXPOSURE]: The skill is designed to read employee-specific data from local files like .agents/reports/[name].md. While these files are sensitive, the skill does not exfiltrate this information or use network tools.
[INDIRECT_PROMPT_INJECTION]: The skill reads untrusted data from local report files without explicit boundary markers, creating a surface for indirect prompt injection. However, the risk is negligible as the agent has no dangerous capabilities (like shell or network access) to exploit. Ingestion points: .agents/reports/ (SKILL.md). Boundary markers: Absent. Capability inventory: None. Sanitization: Absent.

performance-reviews