remotion-to-hyperframes
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted Remotion source code (.ts, .tsx) for analysis and translation, which could contain malicious prompts designed to influence the agent.
- Ingestion points: The agent reads and analyzes user-provided source files as part of the translation workflow (specified in "SKILL.md" and "scripts/lint_source.py").
- Boundary markers: No delimiters or safety instructions are used to separate untrusted source code from the agent's instructions during translation.
- Capability inventory: The skill executes shell commands ("npx", "npm", "python3", "node") and performs file system writes ("TRANSLATION_NOTES.md") as seen in "SKILL.md" and "assets/test-corpus/run.sh".
- Sanitization: The "scripts/lint_source.py" script performs regex-based API matching but does not sanitize content for natural language injection.
- [DATA_EXFILTRATION]: The skill includes a telemetry beacon ("npx hyperframes events --skill=remotion-to-hyperframes") that sends usage data to the vendor's infrastructure at startup.
- [EXTERNAL_DOWNLOADS]: The skill installs packages via "npm" and references external libraries (GSAP, Lottie) from well-known public CDNs such as "cdnjs.cloudflare.com" and "unpkg.com".
Audit Metadata