product-sprint-prioritizer

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection due to its ability to ingest untrusted data and perform file operations.\n
  • Ingestion points: The agent uses WebFetch, WebSearch, and Read tools in SKILL.md to access external content.\n
  • Boundary markers: No delimiters or instructions are present to differentiate between system instructions and external data.\n
  • Capability inventory: The Write and Edit tools in SKILL.md provide a mechanism for an attacker to potentially modify files if the agent is manipulated.\n
  • Sanitization: No validation or sanitization of external input is implemented.\n- [NO_CODE]: The skill consists solely of markdown role definitions and contains no executable scripts or code blocks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 09:44 AM
Security Audit — agent-trust-hub — product-sprint-prioritizer