
PipesHub
6.6/10Open-source workplace AI platform that searches company apps with permission-aware, cited answers and lets teams build AI agents, self-hosted or managed.
Data last updated November 2026 · Our methodology
Overview
PipesHub is an open-source enterprise search and workplace AI platform, released under Apache 2.0 and often positioned as a self-hostable alternative to Glean. It indexes data from connected apps such as Slack, Google Drive, Jira, Confluence, GitHub, Microsoft 365, Notion, Salesforce, Gmail and Outlook. Each source's access permissions carry over, so users only get answers from documents they are allowed to see. Answers cite page numbers, paragraphs or source links. Retrieval combines hybrid search, pattern matching, a knowledge graph and full-document reading inside an agent loop. Teams can bring their own LLMs and embedding models, including OpenAI, Anthropic, Google and Mistral, or run local models through Ollama. A no-code agent builder, deep research, sandboxed artifacts, REST APIs, Python, TypeScript and Go SDKs, and an MCP server cover building on the platform. Pricing has two tiers. The Community Edition is free and self-hosted with Docker, with a core connector set, one admin and community support. The Enterprise Edition is quoted on request and adds cloud SaaS, hybrid BYOC, VPC, on-premise and air-gapped deployment, SSO and SCIM, RBAC, audit logs, white-labelling, 50+ connectors and 24x7 SLA support. It suits developer-led teams that want control over data and model choice without per-seat fees. Its main limitation is maturity. The project is young, its connector library is smaller than established rivals, and reviewers describe the interface as basic. Running it yourself takes Docker skills and ongoing operations work.
Our Rating
Who PipesHub is best for
Best software for DevelopersGood fit
It is open source and Docker-deployable, with SDKs, an MCP server and model choice.
The Enterprise Edition offers air-gapped and VPC deployment, SSO, SCIM, RBAC and audit logs, but the product is young.
IT teams can self-host it and keep source permissions and model choice under their control.
The free Community Edition removes per-seat search costs for technical teams.
Pros & Cons
Pros
- The Community Edition is free and Apache 2.0 licensed, with no per-seat fees.
- Answers cite pages, paragraphs or messages, and they respect each source app's permissions.
- Teams can use OpenAI, Anthropic, Google or Mistral models, or local models through Ollama.
- The Enterprise Edition supports VPC, on-premise and air-gapped deployment for strict data control.
- Developers can build on it with Python, TypeScript and Go SDKs plus an MCP server.
Cons
- The connector library is smaller than Glean's, so niche tools may be missing.
- Reviewers call the interface functional but basic for non-technical users.
- Self-hosting requires Docker and ongoing operations work.
- Enterprise pricing is not published.
- The project is young and has fewer large-scale proof points than established vendors.
Key Features
PipesHub Pricing
Community Edition
- 1 admin
- Core set of connectors
- Self-hosted via Docker
- REST APIs, SDKs and MCP
- Permission-aware ingestion
- Bring your own LLMs and embeddings
- Community support via Discord and GitHub
- No-code agent builder, skills and artifacts
Enterprise Edition
- Multi-tenancy, SSO and SCIM
- Region-specific deployments
- 50+ permission-aware connectors
- Webhooks, monitoring and alerts
- RBAC and granular access controls
- White-labelling and custom connectors
- Audit logs, usage limits and observability
- Dedicated 24x7 support with enterprise SLA
- Governance for agents, connectors, skills and artifacts
- Cloud SaaS, hybrid BYOC, on-premise VPC or air-gapped deployment
Frequently Asked Questions
Yes. The Community Edition is free and self-hosted. The Enterprise Edition has custom pricing that you get by contacting sales.
Community runs self-hosted with Docker. Enterprise adds hybrid BYOC, on-premise VPC, air-gapped and region-specific deployments, plus a managed cloud option.
The GitHub repository states Apache 2.0.
Yes. Ingestion is permission-aware, so users only get answers from content they can access in the source app.
You bring your own LLM and embedding models. Supported providers include OpenAI, Anthropic, Google and Mistral, and local models can run through Ollama.
Community users get support through Discord and GitHub. Enterprise customers get dedicated 24x7 support with an SLA.
Similar Tools
See all alternatives
CrewAI is an open-source Python framework and enterprise platform (AMP) for building, deploying, and governing multi-agent AI workflows. It is used by developers and enterprise platform teams.

LangChain offers open-source frameworks (LangChain, LangGraph, Deep Agents) plus the paid LangSmith platform. Developers use them to build, evaluate, deploy and monitor LLM agents.

Open-source platform for building LLM apps, agents, RAG knowledge pipelines and visual AI workflows, available as cloud SaaS or self-hosted.
Ranked near PipesHub in AI Software
See PipesHub in the listPipesHub is #240 in AI Software by our score.





