capacitor-app-upgrade-v5-to-v6

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFE
Full Analysis
  • [DYNAMIC_CONTEXT_INJECTION]: The skill uses a dynamic command to read the local package.json file and log current Capacitor package versions at load time.\n
  • Evidence: SKILL.md contains a command !node -e "..." that filters dependencies for @capacitor/ prefixes.\n
  • Context: This is used to provide the agent with a snapshot of the current environment to ensure the correct migration path is followed.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local filesystem (package.json) which could potentially contain malicious content if the project itself is compromised.\n
  • Ingestion points: package.json is read by a node script in SKILL.md.\n
  • Boundary markers: None present.\n
  • Capability inventory: The skill allows the use of the Bash tool to perform node operations and project modifications (e.g., npm install, npx cap sync).\n
  • Sanitization: The input from package.json is parsed as JSON and filtered for specific keys starting with @capacitor/, providing structural validation and limiting the data surface.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 10:41 AM