Give coding agents long-term memory as a tree of plain files
IHMT stores agent memory as a file tree that coding agents such as Claude Code, Codex and opencode can read and update.
Source
gonzar79 on Hacker News · 2026-10-12 · Project link
Written in our own words. Credit to the original author. Link checked 2026-10-24.
What they did
IHMT stores agent memory as a file tree that coding agents such as Claude Code, Codex and opencode can read and update.
Result (as claimed)
Memory stays in plain files you can inspect and version.
Tools they used
Cost: Pricing depends on the tools you pick; check each official pricing page.
Set it up
Pick a contained repo or task
Start on a branch or side project where mistakes are cheap.
Install and configure
Set up IHMT, Claude Code, Codex, OpenCode. Read what it can run on your machine and limit permissions where possible.
Run it on a real task
Give it a well-scoped issue and let it work. Review the diff like any pull request.
Add guardrails
Keep tests and CI in the loop, and never let an agent push to main or touch production secrets unattended.
Copy the full setup prompt
Paste into your agent and it will walk you through the setup, asking before anything that sends, posts or spends.
You are helping me set up this AI use case end to end in Claude Code. Goal: Give coding agents long-term memory as a tree of plain files What the original author did: IHMT stores agent memory as a file tree that coding agents such as Claude Code, Codex and opencode can read and update. Reported result (as claimed by the author): Memory stays in plain files you can inspect and version. Source: https://news.ycombinator.com/item?id=50058903 Tools they used: IHMT, Claude Code, Codex, OpenCode Category: Dev & ops Please: 1. Ask me 3 short questions about my situation (accounts I have, budget, what I want the output to look like). 2. Propose a step-by-step plan using these tools or cheaper alternatives, with the time and cost of each step. Use official pricing pages; say "check pricing" if unsure. 3. Then do the setup with me one step at a time, waiting for my OK before anything that sends messages, posts publicly, spends money or deletes data. 4. Finish with a checklist to run it again next time.
You are helping me set up this AI use case end to end in Codex. Goal: Give coding agents long-term memory as a tree of plain files What the original author did: IHMT stores agent memory as a file tree that coding agents such as Claude Code, Codex and opencode can read and update. Reported result (as claimed by the author): Memory stays in plain files you can inspect and version. Source: https://news.ycombinator.com/item?id=50058903 Tools they used: IHMT, Claude Code, Codex, OpenCode Category: Dev & ops Please: 1. Ask me 3 short questions about my situation (accounts I have, budget, what I want the output to look like). 2. Propose a step-by-step plan using these tools or cheaper alternatives, with the time and cost of each step. Use official pricing pages; say "check pricing" if unsure. 3. Then do the setup with me one step at a time, waiting for my OK before anything that sends messages, posts publicly, spends money or deletes data. 4. Finish with a checklist to run it again next time.
You are helping me set up this AI use case end to end in Grok Bot. Goal: Give coding agents long-term memory as a tree of plain files What the original author did: IHMT stores agent memory as a file tree that coding agents such as Claude Code, Codex and opencode can read and update. Reported result (as claimed by the author): Memory stays in plain files you can inspect and version. Source: https://news.ycombinator.com/item?id=50058903 Tools they used: IHMT, Claude Code, Codex, OpenCode Category: Dev & ops Please: 1. Ask me 3 short questions about my situation (accounts I have, budget, what I want the output to look like). 2. Propose a step-by-step plan using these tools or cheaper alternatives, with the time and cost of each step. Use official pricing pages; say "check pricing" if unsure. 3. Then do the setup with me one step at a time, waiting for my OK before anything that sends messages, posts publicly, spends money or deletes data. 4. Finish with a checklist to run it again next time.
The Daily 5
Get five new use cases tomorrow morning
Free. Real sources. Unsubscribe any time.
Guides for this use case
How to run multiple coding agents together · How to use Claude Code and Codex together · How to build agent memory and context layers · How to add memory to coding agents
Related use cases
Organize seven coding agents with one AGENTS.md and a folder convention
kunalsahu20_ runs 7 AI coding agents and shared his full setup: folder convention, the AGENTS.md he uses, templates and 5 prompts.
Split a big job across Claude Code, Codex and OpenCode subagents in parallel
pi-cowboy orchestrates background agents for Pi, each in an isolated copy-on-write Git worktree, can run subagents in Claude Code, Codex, OpenCode and more, and consolidates the results.
Watch all your coding agents in one multiplexer and jump to the next one that needs you
Corgi shows a running activity feed for each coding agent, jumps to the next unread agent with F1, and has one launcher for Codex, Claude Code and Pi across multiple accounts with session search, resume and usage tracking.
Have CLAUDE.md import AGENTS.md so Claude and Codex share rules
Dainer_Jun's CLAUDE.md just imports AGENTS.md and routes the work, Codex reads AGENTS.md on its own, and work state lives in files too.