remotion-spec-translator
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates a workflow where the agent ingests
VIDEO_SPEC.md(untrusted external data) to extract parameters for a multi-step orchestration pipeline. This represents a surface where maliciously crafted specifications could attempt to influence the agent's behavior during task delegation or code generation.\n - Ingestion points: The agent is instructed to read and parse
VIDEO_SPEC.mdto extract scene, timing, and animation requirements as described inSKILL.md.\n - Boundary markers: The instructions do not define specific delimiters or defensive prompts to ensure the agent ignores potentially adversarial instructions embedded within the specification content.\n
- Capability inventory: Through orchestration, the agent invokes tools and specialized skills (e.g.,
/remotion-scaffold,/remotion-component-gen) that perform file system modifications and generate executable React/Remotion code.\n - Sanitization: The orchestration logic lacks explicit validation or sanitization steps for the data extracted from the specifications before it is used to drive subsequent agent actions.
Audit Metadata