{"slug": "chat-with-your-documents-using-garudust-agent-no-vector-database-required", "title": "Chat With Your Documents Using Garudust Agent — No Vector Database Required", "summary": "Garudust Agent, a tool that enables Retrieval-Augmented Generation (RAG) for document querying without requiring a vector database or embedding API calls. Instead, it uses SQLite FTS5 with a trigram tokenizer to index documents (PDF, TXT, CSV, etc.) and performs full-text searches, feeding the top matching chunks to an LLM for context. The system supports multiple languages, allows users to ingest, remove, or clear documents, and can handle file attachments from messaging platforms.", "body_md": "Most RAG tutorials start the same way: \"First, install a vector database…\" Then come the embedding models, the chunking strategies, the similarity thresholds. By the time you can ask a question about a PDF, you've deployed three services and written 200 lines of boilerplate.\nGarudust Agent takes a different path. RAG is built in — backed by SQLite FTS5 with a trigram tokenizer. No vector database. No embedding API calls. Drop a PDF (or TXT, CSV, Markdown, JSON) into the conversation and start asking questions in seconds.\nWhen you ingest a document, Garudust:\ntokenize = 'trigram'\nWhen you ask a question, doc_search\nruns a full-text query against the index and feeds the top matching chunks to the LLM as context. That's the whole pipeline — one SQLite file at ~/.garudust/state.db\n.\nThe trigram tokenizer means it works on any language, including Thai, Chinese, and Japanese, without any tokenizer configuration.\nRAG is enabled by default. The only thing you need to configure is which directories the agent is allowed to read from:\n# ~/.garudust/config.yaml\nsecurity:\nallowed_read_paths:\n- /home/you/documents\n- /data/company-docs\nThat's it. If you want to turn RAG off entirely:\ndisabled_toolsets: [rag]\nStart the CLI:\ngarudust\nThen tell the agent to ingest a file:\nYou: ingest /home/you/documents/employee-handbook.pdf\nAgent: Indexed employee-handbook.pdf — 47 chunks ready for search.\nPreview: \"This handbook outlines the policies and procedures for all employees…\"\nNow ask anything:\nYou: What is the remote work policy?\nAgent: According to the employee handbook, remote work is permitted up to 3 days per week\nfor roles that do not require on-site presence. Employees must notify their manager\nat least 24 hours in advance and maintain availability during core hours (10am–4pm).\nYou never call these directly — the agent decides when to use them based on your question. But knowing they exist helps you understand what's happening.\nIf a document changes, just ingest it again. The old index for that path is replaced automatically.\nYou: Remove the Q1 report from the index.\nAgent: Document removed from index.\nOr clear everything:\nYou: Clear all indexed documents.\nAgent: Removed 5 document(s) from index.\ndoc_search\nsupports full FTS5 syntax, which the agent uses automatically when your question benefits from it:\nYou don't need to write FTS5 queries yourself — the agent figures this out. But if you want to guide it:\nYou: Search for \"termination clause\" in the contract documents.\nIngest your onboarding docs, SOPs, and internal wikis. New team members can ask questions in plain language instead of searching through Confluence.\nYou: ingest /docs/sop-release-process.md\nYou: What approvals are needed before a hotfix can go to production?\nYou: ingest /legal/vendor-agreement-2025.pdf\nYou: Does this contract include a limitation of liability clause? What is the cap?\nIngest a log file and ask questions without writing grep patterns:\nYou: ingest /var/log/app/error.log\nYou: Which service caused the most errors in the last hour?\nYou: Are there any database connection timeouts?\nYou: ingest /project/docs/api-reference.md\nYou: What parameters does the /auth/refresh endpoint accept?\nIf you're running garudust-server\nwith a messaging platform, users can send files directly to the bot. Attachments are automatically saved to a temporary path and can be ingested on request:\nUser sends: quarterly-report.pdf (attached)\nAgent: I received your file. Would you like me to index it for search?\nUser: yes\nAgent: Indexed quarterly-report.pdf — 83 chunks ready.\nPreview: \"Q1 2025 Financial Summary — Total Revenue: $4.2M…\"\nUser: What was the gross margin for Q1?\nAgent: According to the report, gross margin for Q1 2025 was 61.3%,\nup from 58.9% in Q4 2024.\nPlatform attachments (files from Telegram, LINE, Discord, etc.) are always allowed regardless of\nallowed_read_paths\n, since they're written to/tmp/garudust_*\nby the platform adapter.\nYou can ingest multiple files and search across all of them in the same session:\nYou: ingest /docs/policy-2024.pdf\nYou: ingest /docs/policy-2025.pdf\nYou: What changed in the travel expense policy between 2024 and 2025?\nThe agent searches both documents and synthesizes the differences.\nCheck what's indexed at any time:\nYou: What documents have you indexed?\nAgent: 2 documents indexed:\n- policy-2024.pdf | 34 chunks | ingested 2025-05-21 09:14\n- policy-2025.pdf | 38 chunks | ingested 2025-05-21 09:15\nstate.db\n, but searches are scoped to the current conversation key. Starting a new session means re-ingesting if you want to query the same files.Garudust's RAG won't replace a purpose-built vector search pipeline for large-scale production retrieval. But for a developer who wants to ask questions about their documents right now — without running a second service — it's the fastest path from PDF to answer.", "url": "https://wpnews.pro/news/chat-with-your-documents-using-garudust-agent-no-vector-database-required", "canonical_source": "https://dev.to/garudust/chat-with-your-documents-using-garudust-agent-no-vector-database-required-1m61", "published_at": "2026-05-21 06:47:49+00:00", "updated_at": "2026-05-21 07:02:36.238464+00:00", "lang": "en", "topics": ["developer-tools", "large-language-models", "artificial-intelligence", "data"], "entities": ["Garudust Agent", "SQLite FTS5", "Garudust"], "alternates": {"html": "https://wpnews.pro/news/chat-with-your-documents-using-garudust-agent-no-vector-database-required", "markdown": "https://wpnews.pro/news/chat-with-your-documents-using-garudust-agent-no-vector-database-required.md", "text": "https://wpnews.pro/news/chat-with-your-documents-using-garudust-agent-no-vector-database-required.txt", "jsonld": "https://wpnews.pro/news/chat-with-your-documents-using-garudust-agent-no-vector-database-required.jsonld"}}