ai-ppt-powerpoint-generator
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides template
curlandnpxcommands in the documentation. These allow users to manually interact with thehttps://agent.deepnlp.orgAPI. These are standard implementation examples and do not execute automatically. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data via the
promptandbrief_inputsparameters to generate visual content. While this creates a potential surface for indirect injection, it is the primary intended function of the generator and follows standard usage patterns. - Ingestion points: User-supplied
promptanddesign_config.brief_inputsin the API request body. - Boundary markers: Not present in the example payloads.
- Capability inventory: Network operations via
curlandnpx(SKILL.md). - Sanitization: None documented; the skill relies on the backend API for content filtering.
- [CREDENTIALS_SAFE]: The skill requires an API key (
DEEPNLP_ONEKEY_ROUTER_ACCESS) provided via an environment variable. The documentation correctly instructs users to use a placeholder, which is a best practice for secret management.
Audit Metadata