tam-sam-som-calculator

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted data from user-provided marketing materials and sales decks as described in Step 0 of SKILL.md. This creates a surface for indirect prompt injection where malicious instructions could be embedded in the context.
  • Ingestion points: SKILL.md (Gather Context).
  • Boundary markers: No explicit markers or delimiters are used to separate user context from instructions.
  • Capability inventory: The skill can execute a local helper script (scripts/market-sizing.py) for calculations.
  • Sanitization: No sanitization or filtering of the user-provided text is performed.
  • [COMMAND_EXECUTION]: The skill includes a deterministic Python script, scripts/market-sizing.py, which is executed to calculate market estimates. The script uses standard libraries (argparse, sys) and performs basic math without network or sensitive file system access.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 12:54 AM