alterlab-cdm-vfx-pipeline
Pass
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it is designed to ingest and process untrusted data from multiple sources. Ingestion points: The agent is instructed to "search the web for current data" and "read existing project files for context," including screenplays and storyboards (SKILL.md). Boundary markers: The instructions lack any requirement for delimiters or specific warnings to ignore embedded instructions in retrieved data (Absent). Capability inventory: The skill has the ability to read local files, perform web searches, and write deliverables back to the file system (SKILL.md). Sanitization: There is no protocol for validating or filtering external content before it is incorporated into the agent's reasoning or output files (Absent).
- [DATA_EXFILTRATION]: The skill's instruction to "read existing project files for context" (SKILL.md) poses a risk of data exposure. If sensitive files such as configuration files, environment variables, or private keys are stored within the project directory, the agent may inadvertently access and process them, which could lead to unauthorized exposure of credentials or private information.
Audit Metadata