cinefoundry-short-video
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local commands including
node dist/cli.js,curl, andpgrep. These are used for the primary function of the skill (video generation) and system health monitoring. The health checks target127.0.0.1and are considered safe system operations. - [INDIRECT_PROMPT_INJECTION]: The skill accepts a user-defined "topic" which is then passed as an argument to a shell command (
node dist/cli.js short '<topic>'). This creates a vulnerability surface where malicious user input could attempt to escape the command context. - Ingestion points: User-supplied topic in the
SKILL.mdworkflow step 3. - Boundary markers: The instruction uses single quotes around the topic placeholder, which is a minimal boundary marker.
- Capability inventory: Shell command execution via
nodefor video generation andcinefoundryfor memory management. - Sanitization: No mention of input sanitization or escaping, relying on the agent to correctly handle potentially malicious shell metacharacters within the topic string.
Audit Metadata