workleap-squide
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation suggests using the
degittool to fetch application templates from the official Workleap GitHub repository (https://github.com/workleap/wl-squide). This is a standard developer scaffolding practice and targets a trusted vendor repository. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the fetching of global application data via
usePublicDataQueriesanduseProtectedDataQueries. While this introduces an external data ingestion surface where the agent might process API responses, the context is limited to standard React application development, and no high-risk capabilities (like arbitrary command execution) are exposed to the agent. - Ingestion points: Global data fetching hooks in
references/hooks-api.md. - Boundary markers: None explicitly mentioned in the UI implementation examples.
- Capability inventory: Data fetching, routing, and React component registration.
- Sanitization: Standard React DOM rendering sanitization.
- [SAFE]: No other security concerns were identified across the remaining threat categories, including prompt injection, data exfiltration, obfuscation, privilege escalation, and persistence. All code examples represent legitimate usage of the Squide framework.
Audit Metadata