capacitor-app-upgrades
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The skill utilizes dynamic context injection to run node and find commands at load time. These are used strictly to gather project metadata (package versions and file locations) required for the upgrade workflow.
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill guides the user through npm install and npx cap sync, which are standard development operations from verified registries and do not involve untrusted remote scripts.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface through the ingestion of external file content. (1) Ingestion points: Capacitor package names and versions extracted from package.json in SKILL.md via dynamic context. (2) Boundary markers: The content is displayed under the Live Project Snapshot section without specific delimiters. (3) Capability inventory: The skill has access to the Bash tool for performing project maintenance and build commands. (4) Sanitization: The extraction script performs basic validation by filtering keys to only include those starting with the @capacitor/ prefix.
- [SAFE]: No malicious obfuscation, credential harvesting, or unauthorized persistence mechanisms were detected. The skill's behavior is consistent with its stated objective.
Audit Metadata