Khoj
Open-source personal-AI second brain with custom agents, automations, deep research, and a code sandbox.
Khoj is an open-source personal AI, often described as a self-hostable second brain. It lets a person chat with any local or online LLM, get grounded answers from the web and from their own documents (PDFs, Markdown, Notion, Word, org-mode, and more), and reach it from the browser, desktop, phone, Obsidian, Emacs, or WhatsApp.
Beyond retrieval and semantic search, Khoj is genuinely agentic: users can build custom agents with their own knowledge, persona, chat model, and tools, schedule automations for recurring research and notifications, run an autonomous deep-research mode, and execute code in a sandbox. It is model-agnostic across providers such as GPT, Claude, Gemini, Llama, Qwen, and Mistral.
Importantly, the product's footprint has changed. Khoj Cloud, the hosted service at app.khoj.dev, was sunset on April 15, 2026. The project itself remains fully open-source under the AGPL-3.0 license and self-hostable, with active development continuing on GitHub. The team behind it, Khoj AI (a Y Combinator company), has shifted its focus to two newer products: Pipali, an open-source AI coworker that runs locally and can read and write files, use the browser, run code in a sandbox, and integrate with tools like Jira, Linear, and Slack over MCP; and Open Paper, a research assistant for literature review.
For teams evaluating it today, Khoj is best understood as a mature, self-hostable open-source personal-AI project rather than a managed cloud service, with the company's newest energy going into Pipali and Open Paper.
Vendor details
Canonical URL
https://khoj.dev
Category
Agent builder
Company status
independent
Use cases & customers
In practice
You want to ask questions across your own notes and documents without handing them to a cloud vendor. Khoj runs self-hosted and gives grounded answers from your PDFs, Markdown, and Notion alongside the web.
Your assistant only lives in one app. Khoj reaches you from the browser, desktop, phone, Obsidian, Emacs, or WhatsApp, and you can build custom agents with their own knowledge and tools.
You have recurring research you keep doing by hand. Khoj schedules automations for repeat research and notifications and can run an autonomous deep-research mode over your sources.
Agentic Index coverage score
8.0 / 14 capabilities · 57%
| Integrations & Tool CallingEquips agents with tools such as web search and a code execution sandbox and ingests documents from sources including Notion, GitHub repositories, and local files, real tool calling and data connections, though not a broad software as a service connector library. | Partial |
|---|---|
| Workflow OrchestrationProvides an agent engine with a planner and tools plus scheduled automations and deep research that plan, browse, synthesize, and deliver results, real orchestration, though autonomous multi agent teamwork is not documented. | Partial |
| Knowledge Grounding & RAGRuns retrieval augmented generation over your own documents including portable document format, Markdown, Word, and Notion files using semantic vector search and returns verifiable citations, deep knowledge grounding as the core capability. | Full |
| Human Oversight & GuardrailsRuns agent code inside a sandboxed environment, lets slash commands restrict where the assistant looks to avoid hallucination, and provides verifiable inspectable citations, real guardrails, though not a full approval oversight workflow. | Partial |
| Security, Identity & GovernanceSelf hosting keeps data inside your own network with private internal services, multi user authentication, and an administrative panel, real privacy and access control, though enterprise certifications and single sign on are not documented. | Partial |
| Observability & AuditabilityBehavior is inspectable with verifiable citations for every answer and an administrative panel to review settings, real auditability, though not a comprehensive agent tracing and analytics platform. | Partial |
| Memory & State PersistenceAgents keep persistent knowledge bases and the assistant stays stateful and knowledge indexed across sessions rather than starting empty, real state persistence, though a dedicated episodic memory layer is not documented. | Partial |
| Deployment & Data ResidencyIs open source under the Affero General Public License and fully self hostable through pip, Docker, or a virtual server with local models via Ollama for offline use, and its hosted cloud is being deprecated in favor of self hosting, full deployment and data residency control. | Full |
| Prebuilt Agents, Templates & PacksLets users build custom agents with their own persona, knowledge, and tools, but a browsable marketplace or library of prebuilt agents and templates is not documented. | Unable to verify |
| Triggers & Channel CoverageSchedules background automations on a cadence and is reachable across browser, desktop, phone, Obsidian, Emacs, and WhatsApp with delivery to email and webhooks, broad channels and scheduled triggers, though wider external event triggers are less documented. | Partial |
| Model Flexibility & RoutingIs model agnostic and routes across any local or online large language model including those from OpenAI, Anthropic, Google, Meta, Mistral, and DeepSeek, with a chat model selectable per agent, genuine model flexibility. | Full |
| APIs, SDKs & MCP ExtensibilityExposes representational state transfer and websocket application programming interfaces, integrates the Model Context Protocol for tools, and ships as a fully open codebase teams can extend, comprehensive extensibility. | Full |
| Testing, Debugging & OptimizationA dedicated testing, debugging, or evaluation engine for its agents is not documented. | Unable to verify |
| Browser & Computer UseA code execution sandbox lets agents run Python safely, a real form of computer use, though interactive control of a browser is not documented and web access is retrieval only. | Partial |
The Agentic Index coverage score grades every vendor Full, Partial or Unable to verify against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
Free and open source (self-host)
Related vendors
- Lindy — No-code AI agent builder focused on automating repetitive business…
- Activepieces — Open-source automation platform with a self-hosted Community Edition…
- Adopt AI — No code agent builder that turns existing applications and manual…
- Agent.ai — A marketplace and network of hundreds of prebuilt AI agents plus a…
- AgentX — No code platform to build, evaluate, and deploy multi agent…
- AI Library — No code enterprise AI platform to build and run AI agents across the…
Alternatives to Khoj
The closest documented capability profiles to Khoj among agent builders tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- ReN36.0 / 14Fuller documented coverage on Security, Identity & Governance
- Dify9.5 / 14Adds documented Prebuilt Agents, Templates & Packs and Testing, Debugging & Optimization
- Moterra6.5 / 14Adds documented Prebuilt Agents, Templates & Packs
- StackBlitz Bolt8.5 / 14Adds documented Prebuilt Agents, Templates & Packs and Testing, Debugging & Optimization
- AgentX10.0 / 14Adds documented Prebuilt Agents, Templates & Packs and Testing, Debugging & Optimization
- AI Library8.0 / 14Adds documented Prebuilt Agents, Templates & Packs and Testing, Debugging & Optimization
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded