AI Security Review
scanned 2h ago · by lpm-firewall-aiReview flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.
Decision evidence
public snapshot- `bin/chatgpt-web-opencode.js` invokes setup on explicit CLI use.
- `dist/cli/setup.js` writes OpenCode global/project config and adds this plugin.
- `dist/cli/setup.js` renames a stale package cache entry.
- `dist/transport/http.js` spawns its bundled Node fetch helper under Bun.
- `package.json` has no preinstall/install/postinstall hook.
- Config mutation is user-invoked, not import-time or lifecycle-triggered.
- OAuth tokens and conversation data are sent only to OpenAI/ChatGPT endpoints.
- Filesystem writes are limited to OpenCode config and a package-owned device-ID cache.
- No eval, dynamic code loading, shell payload, credential harvesting, or third-party exfiltration found.
Source & flagged code
4 flagged · loading source`bin/chatgpt-web-opencode.js` invokes setup on explicit CLI use.
bin/chatgpt-web-opencode.jsView on unpkg`dist/cli/setup.js` writes OpenCode global/project config and adds this plugin.
dist/cli/setup.jsView on unpkg`dist/cli/setup.js` renames a stale package cache entry.
dist/cli/setup.jsView on unpkg`dist/transport/http.js` spawns its bundled Node fetch helper under Bun.
dist/transport/http.jsView on unpkg