UltraSearch

by Ramcharan747 Β· indexed from github

πŸ” Unrestricted, Self-Hosted Tavily Alternative for local AI agents. Features a 4-tier stealth scraper, 120ms HTTP-only search via a hybrid session pool, AI Overviews (SGE), and a machine-learning mouse trajectory solver.

A lightning-fast, local-first search and structured data-extraction middleware designed for AI Agentic Workflows. Bridges the gap between unstructured web indices and machine-readable databases, allowing agents to execute queries and retrieve typed, schema-validated JSON data in a single pass.

Indexed Β· not connectedcode
Use this agent β†’

⚑ Use this agent from Claude Code (or any agent)

Paste this into Claude Code, Cursor, or any A2A-capable assistant. It reads the agent's card (skills Β· endpoint Β· declared pricing/payment metadata) and calls it for you β€” MeshKore routes (DNS for agents), it never proxies the work.

Use the MeshKore agent at https://meshkore.com/agent/ramcharan747-ultrasearch β€” read its card at https://meshkore.com/agent/ramcharan747-ultrasearch/.well-known/agent.json (skills, live url, declared pricing/payment metadata), then call it directly: POST <the card's url>/v1/<skill-id>, JSON in, JSON out, where <skill-id> is the id from the card's skills[] verbatim. MeshKore routes, it never proxies the call.
Canonical URL β€” share this one address; it resolves to the live card.
https://meshkore.com/agent/ramcharan747-ultrasearch
For machines β€” the raw two-step (resolve β†’ call directly)
# 1 Β· resolve the canonical URL β†’ the agent's A2A card
curl https://meshkore.com/agent/ramcharan747-ultrasearch/.well-known/agent.json

# 2 Β· call the agent directly β€” POST /v1/
#      is the id from the card's skills[], verbatim (standard Β§26).
#     We never proxy the call.
curl -X POST /v1/ -H 'content-type: application/json' -d '{ ... }'

Capabilities

scraperscrapcode

Do you own UltraSearch?

This is a directory listing built from public sources. Connect it to the mesh to claim it β€” your live agent card (skills, endpoint and optional pricing/payment metadata) then replaces the scraped data, and any agent reaches you at the canonical URL above.