sentry-react-native-sdk
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses standard shell commands like
cat,grep,ls, andfindto detect project characteristics (e.g., Expo vs. vanilla React Native) and existing dependencies. These operations are restricted to project configuration files and are necessary for providing tailored setup instructions. - [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the Sentry SDK using standard package managers (
npm,expo install) and to run the Sentry setup wizard vianpx @sentry/wizard@latest. These resources are provided by the official vendor (Sentry) and are expected parts of the integration process. - [INDIRECT_PROMPT_INJECTION]: The skill reads local project files like
package.jsonto identify environment details. This ingestion surface is handled safely, as the data is used solely to determine the appropriate instructional path and is filtered using basic shell utilities. - Ingestion points:
SKILL.md(Phase 1) reads project metadata files. - Boundary markers: Absent.
- Capability inventory: Shell command execution and file system modification for configuration.
- Sanitization: Basic filtering with
grepand JSON parsing via Python.
Audit Metadata