$ loading_
Local-first agentic RAG for searching and accessing team document libraries.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "pharos" yet — see the docs or source repo.
Using Pharos, search our team document library for documents related to the product release process, then summarize the key steps and owners.
Relevant document results plus a concise summary grounded in the retrieved content.
Use Pharos to retrieve information across multiple document formats in the team library and answer: What required steps must a new employee complete during onboarding? Include the supporting sources.
An answer with source-backed references from the retrieved documents.
Use Pharos as an MCP tool to search the team document library for content about customer data access permissions and return only the information allowed by access controls.
Retrieved results and a summary filtered according to access control rules.
Developers, product managers, or researchers can search the team document library and answer questions or summarize findings based on retrieved content. It fits situations where reliable information must be found quickly across scattered documents.
When team documents require permission management, enterprise ACL can limit what is visible during retrieval and answering. This is useful in organizations with strict access boundaries for internal knowledge.
When document retrieval needs to be integrated into an app or AI assistant, the tool can be used through either HTTP or MCP exits. It suits retrieval-augmented workflows built on a team knowledge library.
Pharos is a local-first agentic RAG tool for a team’s document library. It emphasizes multi-format ingest, hybrid retrieval, enterprise ACL, and dual HTTP plus MCP exits.
Based on the provided description, it supports multi-format ingest and hybrid retrieval for team document search and retrieval-augmented Q&A. For the exact supported formats, see the source repository.
The description says it offers both HTTP and MCP exits, so it can be used as a service endpoint or connected as an MCP tool. For exact integration steps, see the source repository.
Generate real Office documents and source code from natural language prompts.
Enable AI agents to transact, check prices, and pay conditionally on Pharos.
Analyze smart contracts before transactions to detect upgradeability and honeypot risks.
Enable AI agents to manage payments on the Pharos testnet.
Set up a local RAG server for private knowledge search and QA.
Assess on-chain transaction risk and provide verdicts with bounded execution plans.