stream-react-native
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the vendor's official CLI (
getstream) and standard development tools (npx,npm,git) to initialize projects, manage development environments, and query SDK documentation paths. - [EXTERNAL_DOWNLOADS]: Fetches SDK migration guides and documentation manifests from GetStream's official GitHub repositories and domains. These resources are required for ensuring the agent applies version-appropriate API patterns and logic.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it ingests untrusted data from local project files (e.g.,
package.json,.env) and external documentation sources during runtime to inform its integration logic. - Ingestion points: Reads project manifests in
SKILL.mdand fetches remote Markdown files for versioning logic inchat-versions.md. - Boundary markers: Absent; the content is treated as authoritative instructional data.
- Capability inventory: Includes file system writes, network requests via
curl, and executing administrative app-wide configuration commands via thegetstreamAPI. - Sanitization: No specific sanitization or filtering of the ingested external or project content is performed before processing.
Audit Metadata