Install Sequna in Claude Code, Cursor, Codex, Windsurf, or any MCP-compatible agent. It captures meaningful progress, organizes work under durable goals, auto-runs safe inbox items, and gives you a structured daily review.
Hi everyone — I built Sequna because I kept running into the same problem with AI coding agents: they were great at helping with the current task, but terrible at preserving the actual work thread across sessions.
Meaningful progress got buried in chats. Todos were scattered. End-of-day review was basically manual reconstruction.
Sequna is my attempt to fix that. It gives AI agents a work memory layer:
- capture meaningful progress
- organize work under durable goals
- auto-run safe inbox items
- generate a structured daily review
It is designed for people already using Claude Code, Cursor, Codex, Windsurf, or other MCP-compatible agents in real workflows.
What I most want feedback on:
1. Is “work memory” a clear way to describe the product?
2. Which part feels most valuable: progress capture, goal organization, inbox auto-run, or daily review?
3. Where does the current setup still feel too heavy?
Report
This feels relevant for anyone who uses AI agents for real project work instead of one-off prompts. Once you’re juggling goals, follow-ups, and unfinished threads, you need something more structured than chat history.
Report
Maker
@sherychen That’s exactly the problem I wanted to solve.
For one-off prompts, chat history is usually enough. But once the work stretches across multiple sessions, goals, and follow-ups, the thread starts to break down fast. Sequna is built for that longer-running mode of working, where the agent needs to retain meaningful progress instead of just remembering the current conversation.
I’m especially interested in whether that framing resonates: not “better chat history,” but a real work memory layer for agent-driven workflows.
Report
No reviews yetBe the first to leave a review for Sequna
This feels relevant for anyone who uses AI agents for real project work instead of one-off prompts. Once you’re juggling goals, follow-ups, and unfinished threads, you need something more structured than chat history.
@sherychen That’s exactly the problem I wanted to solve.
For one-off prompts, chat history is usually enough. But once the work stretches across multiple sessions, goals, and follow-ups, the thread starts to break down fast. Sequna is built for that longer-running mode of working, where the agent needs to retain meaningful progress instead of just remembering the current conversation.
I’m especially interested in whether that framing resonates: not “better chat history,” but a real work memory layer for agent-driven workflows.