sentry-sdk-upgrade

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local shell commands such as cat, grep, and find to identify installed Sentry packages and detect deprecated API patterns within the project's source code.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download and installation of updated Sentry SDK packages using standard package managers (npm, yarn, pnpm). All referenced external documentation and repositories point to official Sentry domains (sentry.io, docs.sentry.io) and the official getsentry GitHub organization, which are trusted vendor resources.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it processes project files (e.g., package.json, source code) which could contain malicious instructions.
  • Ingestion points: package.json and project source files (e.g., *.ts, *.js, *.tsx).
  • Boundary markers: No specific delimiters or boundary instructions are utilized when reading file content.
  • Capability inventory: The skill possesses capabilities to read files, perform pattern matching, and execute package management commands.
  • Sanitization: No explicit sanitization or filtering of file content is performed prior to the AI processing the code logic for migration advice.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 05:40 AM