Self-Hosted Business

As of June 2026, AIDiveForge tracks 11 self-hosted business. Curated self-hosted business tracked by AIDiveForge. Listings are verified against each tool's live website and re-checked regularly.

Last updated June 12, 2026 · 11 tools

1. 

   1. AI-Blueprint

   The repo describes a self-hosted, open-source workspace covering the core legal workflow loop: document-grounded chat with source references, contract review with clause analysis, legal drafting, and matter preparation. Because the whole stack runs locally via Docker, there is no API call carrying privileged documents to a third-party cloud. That tradeoff has a cost — setup requires someone comfortable with Docker, environment files, and database migrations, and there is precious little polish compared to hosted competitors. Teams without an in-house developer will hit the configuration wall before they hit a legal task.

   FreeOpen Source
2. 

   2. Basedash MCP Connectors

   Basedash is an AI-native BI platform where you describe what you want in plain English and it writes the SQL, runs the query, and assembles the dashboard. The vendor states it connects to 750+ data sources, so the warehouse you already use plugs in without a migration. Daily briefings ship automatically, which means your morning standup has numbers before anyone opens a laptop. The ceiling shows up when teams need complex, multi-source joins with custom business logic — the AI gets you to 80%, and a human has to close the gap. Teams that outgrow the generated SQL typically layer in a dedicated analytics engineer to audit and harden what Basedash produces.

   PaidFree Trial · 14 days
3. 

   3. BetterCallClaude

   The tool installs as a plugin in Anthropic's Cowork Desktop and routes legal tasks — contract review, case research, document drafting, compliance checks — across 20 specialized agents, each scoped to a specific practice area. It covers all 20 Italian regions plus national law, and the vendor states legal research runs 70% faster based on activity analysis from Italian firms. The privacy architecture is the real differentiator: local LLM processing via Ollama means your matter data stays inside your own environment, which is the compliance baseline Italian professional secrecy rules demand. The ceiling appears when you need tasks that fall outside its pre-built agent scope or require integrations with external systems — there is no API surface, so automation into case management software requires manual steps.

   FreeOpen Source
4. 

   4. Fundamentalio

   The tool pulls fundamentals via yfinance and sends them through OpenAI in either a quick-scan or deep-research mode, so you can screen a watchlist fast or stress-test a single position with more context. Because every analysis is a one-shot OpenAI call, there is no memory between runs — each report starts cold. The Lynch framing is the differentiator: the prompt logic is built around his specific criteria, not generic financial ratios, which means output reads like a philosophy-aligned verdict rather than a data dump. Self-hosted and MIT-licensed, so your API keys and tickers stay off third-party servers. The ceiling is clear: if your process needs portfolio-level comparison, backtesting, or screening across hundreds of tickers in a session, the architecture does not support it.

   FreeOpen Source
5. 

   5. Judicex

   Judicex runs as a local Flask workspace where you ingest official sources and matter files into a SQLite knowledge base, then draft, chat, and run workflow checks against only what you fed it. The LLM answers are bound to that evidence store — the vendor describes this as an 'answer contract that fails closed instead of hallucinating.' You deploy it on your own infrastructure, which means client files never leave your network. The MCP server lets you connect external tools, and JSON workflow packs let you encode firm-specific matter analysis profiles. The ceiling appears when your team grows past a handful of users — multi-tenant auth and SSO are on the roadmap but not yet shipped.

   FreeOpen Source
6. 

   6. LeaseScan by VantagePoint Networks

   LeaseScan accepts a lease document and returns a scored report flagging problematic clauses, jurisdiction-specific compliance issues, and negotiation points — without requiring a lawyer or a law degree to read the output. The one-shot workflow means you upload, pay, and receive a static report; there is no back-and-forth agent loop, no iterative refinement, and no live chat with the analysis. For individual renters reviewing a single agreement before signing, the model fits well. For property managers who need to process dozens of leases against changing local regulations, the per-scan cost structure and report format become friction. Self-hosted deployment is available for organizations that cannot send lease documents to a third-party server.

   Paid
7. 

   7. MetaLens

   The vendor states the platform deploys eight AI agents that scan a Metabase instance, score its health, flag stale and duplicate content, generate governance documentation, and rebuild dashboards for executive reporting — all without requiring a step-by-step human review of each artifact. The free tier produces a health score and summary, which is enough to quantify the damage before committing budget. The paid tiers unlock the agents that actually fix things: documentation generation, catalog building, gap analysis, and dashboard rebuilding. Teams without in-house Metabase expertise are the explicit target; the tool is designed to substitute for governance infrastructure that most analytics teams never built. The self-hosted Metabase path is supported, and the vendor provides an open-source installer script for deployment.

   PaidFree Trial · 14 days
8. 

   8. RedNotebook AI

   The tool runs a Next.js frontend over a FastAPI backend and connects to Trino, DuckDB, and eleven other SQL engines, so analysts working across mixed data infrastructure do not need a different client per engine. AI suggestions surface inside the notebook for SQL generation, chart selection, and data profiling — including PII detection — without sending your schema to a third-party SaaS layer. The NotebookLM-style knowledge layer lets you ask questions grounded in your actual query results rather than a generic model context. That said, the project carries a low star count and three open issues with no merged pull requests, which means production stability depends on how closely your use case matches what the maintainer has tested. Teams hitting edge cases in multi-engine joins or complex profiling jobs will be patching source code themselves.

   FreeOpen Source
9. 

   9. Rev.com

   Rev is a transcription and analysis platform built specifically for legal and law enforcement workflows, converting audio and video files into searchable, speaker-labeled transcripts with AI-assisted review features layered on top. The free tier gives you 45 minutes per month — enough to evaluate accuracy on your evidence type, not enough for production caseloads. At scale, the workflow is file-in, transcript-out, then search and clip; there's no autonomous step where Rev takes action on your behalf. The SmartDepo feature generates deposition memos with page-line citations, which matters when admissibility turns on exact references. Multilingual intake — the docs describe bilingual interview support — addresses a gap that generic transcription tools ignore entirely.

   Paid
10. 

    10. Sofya

    Sofya targets that gap: an AI layer built for healthcare workflows that handles patient intake, structures notes during consultations, and surfaces clinical decision support in real time. The vendor states full HIPAA and LGPD compliance, HL7 and FHIR integration, and self-hosted deployment for organizations that cannot let patient data leave their infrastructure. Where it fits cleanly is high-volume clinical environments already running compatible EHRs — the structured output lands directly into existing systems rather than creating a parallel documentation layer. The ceiling appears in smaller or more specialized clinical settings where the intake and decision-support logic does not map to the tool's pre-built workflows, and the custom pricing model means budget clarity requires a sales conversation before any technical evaluation.

    Paid
11. 

    11. TetherDust

    TetherDust runs inside your infrastructure, connecting MCP servers to your codebase and database documentation so agents generate SQL that can be checked against the actual schema — not guessed. The core workflow chains natural language input through containerized agents that produce SQL, d3.js dashboards, and schema-to-code dependency maps, all inside strict read-only query boundaries. Scheduled reports ship by email or download without exposing write access. RBAC and audit logging are included for teams where data access needs a paper trail. The ceiling appears when you need write operations, or when your branching query logic outgrows what the agent layer can express without custom extensions.

    FreeOpen Source

Listings on this page are sourced and verified by the AIDiveForge data pipeline. AIDiveForge is editorially independent — no money changes hands for inclusion.