Private Deep Search

Performs private, multi-round deep web searches excluding Google/Bing, synthesizes results with citations, and does not retain user data or logs.

This package appears coherent and implements a local SearXNG + scraper. Before installing: 1) Inspect the Docker image (searxng/searxng:latest) or pin a specific release tag to avoid surprise image changes. 2) Verify docker/searxng/settings.yml after setup (setup.sh replaces the placeholder secret for you). 3) Be aware deep_research.py will fetch and parse URLs returned by searches — consider running the container and scraper in an isolated network namespace or behind a VPN if you want to limit outbound reach (prevents accidental access to internal hosts). 4) Review and, if necessary, adjust IGNORED_DOMAINS and request timeouts/rate limits to avoid aggressive scraping. 5) If you require extra assurance, pull and inspect the searxng Docker image locally or build from source and run with least-privilege container settings (no host network unless you intend VPN=host). Overall: functionally consistent with its stated purpose, but follow standard supply-chain and network-isolation best practices.

Type: OpenClaw Skill Name: privatedeepsearch-melt Version: 1.0.0 The skill bundle is designed for a privacy-first deep research assistant, utilizing a self-hosted SearXNG instance via Docker. The `setup.sh` script correctly initializes the SearXNG configuration by generating a unique secret key and starting the Docker container. The `deep_research.py` script performs web scraping and iterative searches, but all network requests are directed to the local SearXNG instance or legitimate external search engines/websites for content retrieval, with a clear focus on local processing and privacy (e.g., blocking tracking domains). There is no evidence of data exfiltration, malicious execution, persistence mechanisms, prompt injection against the agent, or obfuscation. All actions are transparent and aligned with the stated purpose.