inworld
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied text for speech synthesis and retrieves external documentation from a remote vendor-controlled URL.
- Ingestion points: User input text intended for the
text-to-speechAPI call and remote content fetched fromhttps://www.pixazo.ai/models/inworld.md. - Boundary markers: Absent; the skill does not define specific delimiters for untrusted content.
- Capability inventory: File system write access (
~/.pixazo/api-key), network operations (viacurl,requests, andfetch), and shell command execution (chmod). - Sanitization: Absent; the instructions do not specify validation or filtering of external content before processing.
- [EXTERNAL_DOWNLOADS]: Fetches model configuration and parameter guidelines from the vendor's official domain.
- Evidence: The skill instructs the agent to load
https://www.pixazo.ai/models/inworld.mdfor reference. - [COMMAND_EXECUTION]: Instructs the agent to execute shell commands to manage permissions for the local credential store.
- Evidence:
chmod 600 ~/.pixazo/api-keyused during the API key setup process.
Audit Metadata