pitch-gen

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its handling of user-supplied data in AI prompts.
  • Ingestion points: The argument from the command line is passed directly to the generatePitch function in src/index.ts.
  • Boundary markers: The prompt 'Create pitch deck content for: ${idea}' does not use any delimiters (such as triple quotes or XML tags) or instructions to ignore embedded commands.
  • Capability inventory: The skill uses the fs module to write generated content to the filesystem (src/cli.ts).
  • Sanitization: There is no validation or filtering of the input string to remove potential injection patterns before it is sent to the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 02:05 AM
Security Audit — agent-trust-hub — pitch-gen