alterlab-cdm-film-pitch
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core workflow of ingesting and acting upon untrusted external data.
- Ingestion points: The agent is explicitly instructed to "read existing project files for context" (SKILL.md) and "search the web for current data" (SKILL.md) to gather information for pitch development.
- Boundary markers: There are no instructions defining clear boundaries or delimiters for external data, nor are there guidelines for the agent to ignore potentially malicious instructions embedded within those files or search results.
- Capability inventory: The skill allows the agent to perform web searches and write executable or structured markdown files directly to the project directory.
- Sanitization: The workflow lacks a validation or sanitization step to ensure that instructions found in external data do not override the agent's primary goals during the "self-review" or "crystallization" phases.
Audit Metadata