capacitor-platforms

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run CLI commands for platform management and dependency installation, such as 'npm install', 'npx cap add', and 'npx cap sync'.
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing official '@capawesome/capacitor-electron' and '@capawesome/capacitor-tauri' packages from the npm registry and referencing the 'capacitor-tauri' Rust crate. It also suggests adding hosted MCP servers for documentation access.
  • [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by fetching external documentation through MCP servers. Ingestion points: Tools like 'search_docs' and 'get_doc_page' retrieve data from 'mcp.capawesome.io' and 'capacitor-mcp.capawesome.io'. Boundary markers: No explicit markers or instructions to ignore embedded commands in the documentation are provided. Capability inventory: The agent can perform file writes and execute shell commands to configure the Capacitor environment. Sanitization: No sanitization of the retrieved documentation content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 09:36 AM