# Tempreon — Full Reference > Tempreon is the intelligence layer between you and every AI tool — Claude, ChatGPT, Cursor, and anything MCP-capable. It learns how you think, decide, and work, so every AI tool you use produces output that is yours, not generic. ## What Tempreon is Every AI tool is good on its own. Tempreon is what makes the output *yours*. It is a personal intelligence layer that captures how an individual thinks, decides, writes, and works, then makes that available to every AI tool they use through the Model Context Protocol (MCP). The distinction that matters: memory and portable context are the floor — table stakes that Tempreon includes and does well. The differentiator is the layer above the memory: it learns your judgment, your voice, and your decision patterns, and applies them across every tool and every kind of work. Memory remembers what you said. Tempreon understands what you meant, and works the way you would. ## The problem it solves - Every AI session starts from zero. You re-paste your bio, re-explain your role, re-describe the project — a tax paid every time you open a tool. - Context evaporates between tools. What you told one assistant does not carry to the next. Each tool is an island. - Output sounds like AI, not like you. Technically fine, but your frameworks, tone, and judgment do not come through. Tempreon fixes all three: one identity layer, learned once and refined continuously, reachable from every AI tool you use. ## How it works Your identity, knowledge, decisions, and patterns live in Tempreon. Your AI tools (Claude, ChatGPT, Cursor, and any MCP-capable client) connect to Tempreon as MCP clients and pull what they need at the moment they need it. As you work, Tempreon learns — patterns emerge, instincts form, and the system gets better at anticipating how you think and what you need. The intelligence compounds with use. ## The seven pillars - Core Imprint — your identity, voice, values, and decision patterns, captured once and refined as you work. - The Kernel — the earned half of your identity: versioned operating rules distilled from how you actually work, in your own words, portable to any LLM (https://tempreon.com/features/kernel). - Knowledge Vault — everything you know, structured, searchable, and growing automatically from your conversations. - Tempering & Honing — the learning system: passive learning (Tempering) plus active refinement (Honing) that compound over time. - Decision Gates — your judgment formalized into rules the system learns from how you actually decide. - Agents — autonomous AI personalities you create and dispatch to work in your voice, on your terms. - Bridges — connections to every AI tool you use, so one identity travels across all of them. ## Who it is for Tempreon serves the AI-native operator — one persona, many jobs. The same person writes, codes, decides, plans, learns, sells, and remembers across work, side projects, and home. Documented use cases include executives and leaders, consultants and advisors, solopreneurs and founders, developers, sales professionals, project managers, knowledge workers, content creators, customer support, researchers and analysts, and everyday life. ## Differentiation When everyone has access to the best AI models and prompts, the model is not the differentiator — you are. Tempreon's value is the depth of personalization: your identity, context, and accumulated judgment. It is person-centered and portable: your data is yours, exportable, and reachable from every tool. If you switch AI providers tomorrow, your Tempreon comes with you. Tempreon is the opposite of lock-in. Tempreon is not memory-for-developers. It is for people who want the benefits of a learning system and identity-aware output without building or managing the infrastructure themselves. ## Pricing The free tier is permanent and genuinely useful — not a demo or a trial. Paid tiers (Standard, Premium, Enterprise) go deeper: more learning depth, more scale, and more agents. An early-adopter Founders Circle rate-lock protects early pricing. Current pricing details live at https://tempreon.com/pricing. ## Connect Tempreon runs as an MCP server: URL https://api.tempreon.com/functions/v1/tempreon-mcp/mcp — transport Streamable HTTP, auth OAuth 2.1 with PKCE and dynamic client registration (no API keys). Any AI client that supports adding custom remote MCP servers with OAuth can connect. Dedicated step-by-step setup guides exist for Claude (web/desktop/mobile), Claude Code, ChatGPT, Codex, Gemini, Google Antigravity, Gemini CLI, Cursor, VS Code (GitHub Copilot + Visual Studio), Windsurf, Zed, JetBrains IDEs, Cline, Kiro, Perplexity, Le Chat, Grok, Raycast, Warp, OpenCode, Qwen Code, Kimi, GLM, LM Studio, and open-source clients (Open WebUI, LibreChat, Goose). Each lives under https://tempreon.com/support/docs/bridges — for anything else that speaks MCP, the universal guide at https://tempreon.com/support/docs/bridges/connect-any-client covers the OAuth pattern. ## Page index Every indexable page, for retrieval systems that want the complete catalog alongside the reference prose above. ### Core - [Features overview](https://tempreon.com/features): The seven pillars and how the learning system works. - [Pricing](https://tempreon.com/pricing): Free forever, plus Standard, Premium, and Enterprise tiers. Founders Circle early-adopter rate-lock. - [Use cases](https://tempreon.com/use-cases): Executives, consultants, solopreneurs, developers, sales, project managers, knowledge workers, content creators, support, researchers, and everyday life. - [About](https://tempreon.com/about): Why Tempreon exists. ### Features - [Core Imprint](https://tempreon.com/features/core-imprint): Your identity, voice, values, and decision patterns. - [The Kernel](https://tempreon.com/features/kernel): The earned half of your identity — versioned operating rules from how you actually work, portable to any LLM. - [Knowledge Vault](https://tempreon.com/features/knowledge-vault): Everything you know — structured, searchable, and growing from your conversations. - [Tempering & Honing](https://tempreon.com/features/learning): Passive learning and active refinement that compound with every use. - [Decision Gates](https://tempreon.com/features/decision-gates): Your judgment, formalized into rules learned from how you actually decide. - [Agents](https://tempreon.com/features/agents): Autonomous AI agents that work in your voice, on your terms. - [Bridges](https://tempreon.com/features/bridges): One identity across Claude, ChatGPT, Cursor, and every MCP-capable tool you use. ### Resources - [Blog](https://tempreon.com/blog): Writing on personal AI, individuality at scale, and getting more from the AI you already use. - [Documentation](https://tempreon.com/support/docs): Setup guides and MCP connection instructions. - [Connect any MCP client](https://tempreon.com/support/docs/bridges/connect-any-client): The universal connection flow plus the client compatibility catalog. - [Connection troubleshooting](https://tempreon.com/support/docs/bridges/troubleshooting): Fixes for the common MCP connection failures. - [Glossary](https://tempreon.com/glossary): Tempreon's vocabulary, defined. - [Guides](https://tempreon.com/guides): Long-form practical guides to personal AI, memory, portability, and MCP. - [Security](https://tempreon.com/security): Data security, privacy, and compliance. ### Setup guides (per client) - [Bridges overview](https://tempreon.com/support/docs/bridges): How Bridges work and the full supported-platform table. - [Claude](https://tempreon.com/support/docs/bridges/claude-desktop): Web, desktop, and mobile — every plan, including Free. - [Claude Code](https://tempreon.com/support/docs/bridges/claude-code): Terminal agent. - [ChatGPT](https://tempreon.com/support/docs/bridges/chatgpt): Via developer mode. - [Codex](https://tempreon.com/support/docs/bridges/codex): CLI, IDE extension, and desktop app. - [Gemini](https://tempreon.com/support/docs/bridges/gemini): The Gemini app, as a custom app. - [Google Antigravity](https://tempreon.com/support/docs/bridges/antigravity): Agentic IDE and CLI. - [Gemini CLI](https://tempreon.com/support/docs/bridges/gemini-cli): Terminal agent. - [Cursor](https://tempreon.com/support/docs/bridges/cursor): MCP settings in the editor. - [VS Code](https://tempreon.com/support/docs/bridges/vs-code): GitHub Copilot and Visual Studio. - [Windsurf](https://tempreon.com/support/docs/bridges/windsurf): Including Devin Desktop. - [Zed](https://tempreon.com/support/docs/bridges/zed): Code editor. - [JetBrains IDEs](https://tempreon.com/support/docs/bridges/jetbrains): All JetBrains IDEs plus Junie. - [Cline](https://tempreon.com/support/docs/bridges/cline): VS Code extension. - [Kiro](https://tempreon.com/support/docs/bridges/kiro): AWS agentic IDE. - [Perplexity](https://tempreon.com/support/docs/bridges/perplexity): Multi-model assistant. - [Le Chat](https://tempreon.com/support/docs/bridges/le-chat): Mistral's assistant. - [Grok](https://tempreon.com/support/docs/bridges/grok): xAI's assistant. - [Raycast](https://tempreon.com/support/docs/bridges/raycast): Launcher AI on macOS and iOS. - [Warp](https://tempreon.com/support/docs/bridges/warp): Terminal. - [OpenCode](https://tempreon.com/support/docs/bridges/opencode): Terminal agent, 75+ model providers. - [Qwen Code](https://tempreon.com/support/docs/bridges/qwen-code): Terminal agent. - [Kimi](https://tempreon.com/support/docs/bridges/kimi): Kimi Code CLI via the Claude Code route. - [GLM](https://tempreon.com/support/docs/bridges/glm): Z.ai in Claude Code and more. - [LM Studio](https://tempreon.com/support/docs/bridges/lm-studio): Local models, desktop app. - [Open-source clients](https://tempreon.com/support/docs/bridges/open-source-clients): Open WebUI, LibreChat, Goose, and more.