crashlytics
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill configures an MCP server that downloads the official
firebase-toolsCLI from the NPM registry usingnpx. This is a standard integration pattern with Firebase services (a well-known service provider). - [COMMAND_EXECUTION]: Utilizes
git logandgit blameto identify code owners and contributors for specific crashing files. These commands are legitimate for the stated purpose of developer assignment. - [PROMPT_INJECTION]: Ingests external crash data, including error messages and stack traces from Firebase Crashlytics. While this data is untrusted and boundary markers are absent in the prompt instructions, the skill's capabilities are restricted to local code inspection and report generation, minimizing the risk of indirect prompt injection.
- [SAFE]: No signs of obfuscation, hardcoded credentials, persistence mechanisms, or unauthorized network exfiltration were found. The skill's functions are transparently documented and consistent with its intended use case.
Audit Metadata