action-rhythm-editing
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates external data from variables such as
$action-showcase-direction,$action-choreography-reference, and$seedance-audioto determine action states and timing (found in SKILL.md, references/close-exchange-density.md, and references/rhythm-map.md). - Ingestion points: Data is pulled from external references to set the physical profile and chapter structure of the action sequences.
- Boundary markers: There are no specific delimiters or guardrails instructed to prevent the agent from executing instructions that might be embedded within the data retrieved from these external variables.
- Capability inventory: The skill is restricted to generating natural language prompts and does not utilize tools for file system access, network requests, or shell command execution.
- Sanitization: The instructions do not include steps to sanitize, escape, or validate the content of the external variables before they are interpolated into the final prompt output.
Audit Metadata