algo-hr-compensation

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to its core function of processing external, untrusted datasets.
  • Ingestion points: Untrusted data enters the agent context in SKILL.md (Phase 1), where the agent is instructed to collect internal compensation data and market survey data.
  • Boundary markers: The skill body lacks explicit delimiters or instructions to ignore embedded commands within the ingested HR or survey data.
  • Capability inventory: The skill provides Python reference code in references/pay-equity.md involving data processing libraries, which suggests the agent will perform computations on the ingested data.
  • Sanitization: There are no specific instructions or algorithmic steps provided to sanitize input data or validate it against a schema to prevent instructions embedded in data fields from influencing the agent's behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 02:05 PM
Security Audit — agent-trust-hub — algo-hr-compensation