remotion-engineer
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a potential attack surface where the agent ingests external data from web sources via tools such as BrowserOS. The instructions explicitly address this risk by mandating that external data be treated as hostile and that secrets or PII be redacted before sharing context with subagents. \n
- Ingestion points: External web content via BrowserOS (SKILL.md). \n
- Boundary markers: Explicit instructions are present to treat inputs as hostile and redact sensitive information. \n
- Capability inventory: Terminal execution via
ExecutionProxy(e.g.,rtk npm test) and file-writing capabilities via theMemoryStoreinterface. \n - Sanitization: Instructions for manual redaction of secrets/PII are provided to the agent. \n- [COMMAND_EXECUTION]: The skill references a terminal execution interface (
ExecutionProxy) used to run commands likenpm test. This is utilized as part of a deterministic validation process within the framework's development lifecycle.
Audit Metadata