FilterHN

Ask HN: Has anyone achieved recursive self-improvement with agentic tools?

5 points

9 hours ago

| 4 comments

It feels like all the necessary components are finally available to build a self-reinforcing development loop.

Theoretically, we can now task tools like Claude Code or OpenClaw to monitor a git repo, analyze the abstractions in completed work, and then autonomously generate new agents or skills capable of handling similar tasks in the future.

Is anyone successfully running a loop like this? I’m curious if anyone here has shifted the majority of their time from writing code to crafting these systems—essentially bootstrapping agents that learn from the repo history to build better agents. I'd love to hear from those pushing the boundaries on this.

▲

drsalt

6 hours ago

[-]

I've tried to replicate the real world, so I give my agents backstories, triabl loyalties, and deep-seated character flaws. my agents try to dominate and manipulate each other. they make sure to take credit for every line code. I have manager agents that promote based on shared hobbies. so far it's going well.

▲

acoyfellow

4 hours ago

[-]

I do a fun orchestration system for long running loops on exe.dev (small write up docs.coey.dev) and I feel like I have super powers.

Self healing, I try two ways:

1) use a memory tool to store learnings for next iteration (Deja.coey.dev) and have the loop system instructions tell how to use it. One orchestrator, and sequential worker agents who run til their context is full and then hand off to the next run with learnings

2) the agent Shelley on exe can search past convos when promoted too for continuation.

I’ve been doing this with great success just “falling” into the implementation after proper guardrails are placed

▲

haute_cuisine

2 hours ago

[-]

Any released projects yet?

▲

acoyfellow

2 hours ago

[-]

I release stuff all the time, so yes I suppose? Not trying to spam but you can find a mix of open source, b2b saas i own, and client saas I manage on my personal website.

▲

haute_cuisine

1 hour ago

[-]

I checked your website, it's quite impressive to release 1-3 opensource projects per day.

▲

sdrinf

5 hours ago

[-]

I'm working on something like this. Specifically, I'm doing recursive self-improvement via autocatalysis -but predominantly in writing/research / search tasks. It's very early, but shows some very interesting signs.

The purely code part you described is a bit of an "extra steps" -you can just... vscode open target repo, "claude what does this do, how does it do it, spec it out for me" then paste into claude code for your repo "okay claude implement this". This sidesteps the security issue, the deadly trifecta, and the accumulation of unused cruft.

▲

nycdatasci

9 hours ago

[-]

To head off the semantics debate: I don't mean a model rewriting its own source code. I'm asking about 'process recursion'—systems that analyze completed work to autonomously generate new agents or heuristics for future tasks.

▲

jvalencia

9 hours ago

[-]

-ish. I often keep md files around and after a successful task. I ask Codex to write the important bits down. Then, when I come around to a similar task in the future, I have it start at the md file. It's like context that grows and is very localized. It helps when I'm going through multiple repos at multiple levels.

▲

marc_g

6 hours ago

[-]

I’m also doing similar with fairly decent results. AGENTS.md grows after each session that resulted in worthwhile knowledge that future sessions can take advantage of. At some point I assume it will be too big, then it’s back to the Stone Age for the new agents, in order to release some context for the actual work.