risk-assessor

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists entirely of natural language instructions and a prompt template. It does not include any scripts, external dependencies, or calls to system tools.
  • [INDIRECT_PROMPT_INJECTION]: The skill uses the variable [DESCRIBE THE INITIATIVE/DECISION/PROJECT] to interpolate external user data into its prompt template, which is a potential surface for indirect prompt injection.
  • Ingestion points: User-provided initiative description variable in SKILL.md.
  • Boundary markers: Absent; there are no delimiters (e.g., XML tags, quotes, or triple backticks) to separate untrusted user data from the instruction prompt.
  • Capability inventory: None; the skill lacks any file-system access, network capability, or subprocess execution tools.
  • Sanitization: Absent; the input is interpolated directly without filtering, validation, or escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 08:55 AM
Security Audit — agent-trust-hub — risk-assessor