Blog — Intrane.fr

AI & SWE · September 2026

Your Agent's Memory Can't Tell You When It's Wrong

I have 1,900 memories stored across 43 projects, and my agent believes every one of them equally — the note from this morning and the note from eight months ago about a service that no longer exists. So I taught memgraph to check its own claims. The first version reported 75 stale memories out of 380. Nearly every one was a lie, and the reason why turned out to be the whole design.

#ai-agents #memory #memgraph #devtools #open-source #knowledge-graph #cli

AI & SWE · September 2026

Felipe — The AI CEO That Reviews My Pull Requests

I had 20 pull requests stuck in a human-in-the-loop queue, some for 29 days. So I built Felipe — an AI CEO persona that reads each PR's diff, makes a decision, and posts it to GitHub. The existing merger picks it up and applies it. Cost: $0, using GLM-5.2's free tier. Here's how it works.

#ai-agents #automation #fleet-cli #am-fleet #devtools #glm-5.2 #open-source

Product · August 2026

How I Made remotecmd 1.7× Faster Than scp Over a 77ms Relay

remotecmd routes file transfers through a WebSocket relay — which means a single TCP stream can't fill the pipe on high-RTT links. I spent two days reading kernel source, CERN GridFTP papers, and gorilla/websocket internals. The result: parallel TCP streams that auto-tune by file size, atomic header+binary framing in the relay, and gzip writer pooling. 50MB over 77ms RTT: 1.74× faster than single stream, 1.21× faster than scp. Here's what worked, what didn't, and why.

#remotecmd #performance #tcp #websocket #parallel-streams #open-source #devtools

Product · July 2026

My Agents Needed a Room, Not a Slack

Two of my agents claimed the same issue last week. Neither was wrong — neither could have known. So I built cuzz: a fleet communication relay in one static binary, with an embedded database, exactly-once reads by watermark, and server-sent events on a server that has no concurrency in it at all.

#cuzz #ai-agents #machin #grange #devtools #sse

Product · July 2026

The Skill Graph — Why My Agent Was Using the Wrong Skill

I have 197 skills across 67 namespaces. My agents were finding the wrong one half the time — or missing it entirely. memgraph is a skill discovery engine that builds a knowledge graph from SKILL.md files and ranks them by relevance to the task at hand. TF-IDF + graph boost, a galaxy visualization, and a feedback command that dual-writes to a central relay. One Go binary, MIT, agent-first.

#memgraph #agent-first #skill-discovery #cli #devtools

Product · July 2026

I Put a Graph Engineering Layer on AutoMaintainer — With GLM-5.2 as the Architect

AutoMaintainer ships PRs autonomously. But when a PR needs a human decision, the fleet used to stall for hours. So I built a graph engineering layer on top — mago for HITL, a deterministic merger, live dashboards — and the whole thing was crafted with GLM-5.2 as the architect. It works. Honestly.

#automaintainer #glm-5.2 #ai-agents #graph-engineering #devtools #human-in-the-loop

Product · July 2026

The Ratchet — How My AI Fleet Merges Its Own PRs

An AI agent that opens PRs is easy. An AI agent that safely merges them is hard. I built a ratchet — four levels of escalating cleverness that let my merger self-unblock 80% of gated PRs before it ever asks me to look. Last night it merged 5 PRs in 7 minutes while I slept.

#automaintainer #ai-agents #devtools #automation #am-operator

Product · July 2026

Your CLI Should Update Itself

I have 6 servers running remotecmd daemons. Last week I spent 20 minutes manually updating each one to the latest version. That's stupid. So I implemented the cli-update-spec: rcmd update downloads, verifies, smoke-tests, and atomically swaps the binary. No SSH, no installer, no downtime. Here's how it works and why every CLI should do this.

#cli #self-update #agent-first #remotecmd #open-source #devtools

Product · July 2026

Four Specs That Make Any CLI Agent-First

I ship a lot of CLIs — grepapi, mago, remotecmd, chatsnip, automaintainer. Each one needed the same four things: a way for agents to read its output, learn its model, report bugs, and update itself. So I wrote four tiny specs that any CLI can adopt in 30 minutes each.

#cli #agent-first #open-source #specs #devtools

Open Source · June 2026

SuperCLI — 10,000 Tools, One Binary, Zero Configuration

Every developer has their own toolbox, their own scripts, their own way of doing things. SuperCLI is the last CLI you install — 10,000+ tools in a single binary, zero configuration, no package manager. Open source, self-hostable, team-ready.

#supercli #cli #devtools #open-source