Skip to content

Instantly share code, notes, and snippets.

@danialhasan

danialhasan/founder_memory_stack.md

Created June 15, 2025 00:05
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save danialhasan/fc924db2cee74ea6f57c06af33646553 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save danialhasan/fc924db2cee74ea6f57c06af33646553 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
founder_memory_stack.md

Founder Vertical Memory Protocol

This document defines the vertical memory system for a solo founder operating with AI agents (O3Pro, Claude Code) in your monorepo.

1. Daily Memory Capture

  • New chat per day:

    • Title: YYYY-MM-DD (e.g., 2025-06-11)
    • Import the latest Weekly Control Review table (from canvas or markdown file)

  • End-of-day (EOD):

    • Write an agent_summary block at the bottom:

      ### agent_summary – YYYY-MM-DD
- o3pro-recsys: [short bullet]
- claude: [short bullet]
- (others…)

    • Pin it or mark clearly in the chat/file.

2. Weekly Memory Consolidation

  • Every Monday, 11AM:

    • Open the current chat or markdown file with the canvas/table view.

    • Gather the last 7 days of agent_summary blocks.

    • Paste them into a new section/message.

    • Prompt:

      Run control review for [date range]. Surface all progress, blockers, and next actions in table form.

  • Update:

    • Overwrite the previous Weekly Control Review table with the new one.
    • Refresh meta-summary and checklist based on actual progress.

3. Propagation & Evolution

  • When starting a new daily chat, import the latest table/canvas from the previous session or file.
  • Do NOT depend on chat memory—trust the explicit table/canvas only.
  • Workstreams are updated, added, or pruned in the weekly review.

4. Manual as Default, Automation Optional

  • Steps are fully manual until Stage 2 (CLI/database) automation is implemented.
  • Calendar reminders exist for both daily memory block and weekly review.
  • Protocol must take <10 minutes daily, <30 minutes weekly.

5. Scalability Guideline

  • If agent count or memory management becomes a bottleneck, implement Stage 2 automation:

    • CLI for logging summaries
    • Supabase table for weekly rollup
    • React dashboard for persistent board view

6. Protocol Ownership

  • Only the founder (you) maintain and update the memory protocol and table.
  • AI agents cannot write to canvas/table—only you manage state propagation.

TL;DR

This protocol keeps your vertical memory unbroken, your execution compounding, and your stress low—even as agent complexity grows.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment