fal-lip-sync

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The scripts talking-head.sh and lip-sync.sh utilize curl to interact with external AI services.
  • Evidence: The scripts make POST and GET requests to https://fal.run and https://queue.fal.run to submit jobs and poll for completion.
  • [PROMPT_INJECTION]: A vulnerability for indirect prompt injection (schema confusion) exists in the way text input is handled in talking-head.sh.
  • Ingestion points: The --text argument in scripts/talking-head.sh accepts arbitrary strings that may originate from untrusted external sources processed by the agent.
  • Boundary markers: The script does not utilize boundary markers or 'ignore' instructions when processing this input.
  • Capability inventory: The script has the capability to perform network operations (curl) and file system writes (via .env creation).
  • Sanitization: There is no escaping or validation of the $TEXT variable before it is interpolated into a JSON heredoc. An input containing double quotes could break the JSON structure or inject additional fields into the API request.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 08:00 PM