helmor-debug-operate

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill injects a debug IPC helper that can invoke arbitrary internal Tauri commands and read large terminal/log buffers — a powerful debug backdoor that can be abused to read/exfiltrate sensitive data or trigger backend actions if misused.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The required workflow uses Tauri MCP to read the live Helmor webview DOM/accessibility snapshots (e.g., webview_dom_snapshot / webview_execute_js returning textContent), which can include outsider-authored free text from the app UI (e.g., messages, repo/issue content) and is then ingested into the agent’s LLM context.