Kept

Your AI chats, saved as Markdown locally with no cloud

MacArtificial IntelligenceProductivity

▲ 101 votes19 commentsLaunched May 26, 2026

Visit Website

Daily #20Weekly #28

Kept: is an AI chat and productivity tool on your local archive. Captures conversations from ChatGPT, Claude, Gemini, Grok, and Kimi as Obsidian compatible markdown on your filesystem, with full-text search, knowledge graph, and an MCP servers. MIT license. View your Vault locally,

AI Analysis

📝 Summary

Kept is a local-first AI chat archive tool that automatically captures conversations from ChatGPT, Claude, Gemini, Grok, and Kimi, saving them as Obsidian-compatible Markdown files directly on the user's filesystem. Core features include full-text search, knowledge graph visualization, MCP server support, and local vault viewing with no cloud dependency. It solves key pain points like scattered AI chat histories, privacy risks from cloud services, lack of persistent ownership, and poor searchability of past insights. The unique selling point is its MIT open-source license enabling full user control and seamless integration into personal knowledge management workflows. Overall value proposition: transform ephemeral AI interactions into a private, searchable, personal knowledge base.

📈 Market Timing

In 2025-2026, AI usage is exploding with users generating massive chat data daily, while privacy regulations tighten and local-first software trends rise amid cloud fatigue. Obsidian's growing popularity in PKM perfectly complements local AI archiving needs. Technology for local capture and knowledge graphs is mature. This is an opportune time before major platforms lock down histories further. Excellent Timing.

✅ Feasibility

High. Technical difficulty is moderate: capturing from multiple web AI interfaces is achievable via desktop app or extensions, with local processing keeping complexity manageable. Dev/operation costs are low due to open-source MIT model and no cloud infrastructure. Minimal compliance risks as it's fully local with no data sharing. Strong scalability via community contributions and easy integration with existing tools like Obsidian.

🎯 Target Market

Main segments: Tech-savvy AI power users, knowledge workers, researchers, writers, and Obsidian enthusiasts (ages 25-45) in productivity and software development industries. Geographic focus: Global with concentration in North America, Europe, and Asia (esp. China for Kimi users). TAM for AI productivity tools exceeds $15B, SAM for local PKM archiving ~$800M, SOM for this niche ~$80M. Core pain points: inability to own/search AI-generated insights and privacy concerns. Potential willingness to pay is medium-high for premium extensions, though core product is free.

⚔️ Competition

Medium. Direct competitors: 1. Obsidian Copilot/Smart Connections plugins (obsidian.md), 2. ChatGPT Exporter browser tools (various GitHub projects), 3. Mem.ai (mem.ai), 4. Reflect (reflect.app), 5. Capacities (capacities.io). Advantages vs competitors: fully local/no-cloud privacy, native multi-AI support (5 platforms), built-in knowledge graph, free MIT license, seamless Obsidian compatibility. Disadvantages: potentially less polished interface than commercial cloud tools, relies on user setup for capture, limited marketing as a new PH launch.

Upgrade Pro to unlock full AI analysis