Dev builds aitop, a C++ AI-process monitor, with Claude Code in two hours
johnathonme · reddit · 2026-10-22
Developer johnathonme built aitop — "top for AI workloads" — from scratch using Claude Code running Claude Opus 5.5. It's a 2,500-line C++20 command-line tool with an ncurses UI across six source files, targeting Linux and macOS.
What it does
- Filters AI tools out of the process list and shows CPU, memory and disk I/O, sortable and filterable
- Maps which AI processes own which network connections, plus interface rates
- Shows GPU utilization when drivers are present
- Displays live agent sessions when running greatVibe nodes
- Emits a JSON snapshot with --once, or one JSON line per sample for scripts
- Read-only: never kills processes or prints raw command lines or prompts
How Claude Code was used
The author wrote the brief and steered; Claude Code did the building: the /proc collector, ncurses UI and JSON output; network flow tracking added in v0.2; a macOS collector (libproc, Mach, sysctl) in v0.5; a CTest suite plus a Python test driving the real terminal UI; and CMake builds for Linux x8664/arm64 and macOS Intel/Apple Silicon. The first working version took about two hours including builds, with v0.2–0.5 following over the next day. His takeaway: keep prompts short and correct course with mid-turn steering rather than restarting.
Known gaps: no Apple GPU stats; CPU is per-core so busy processes can exceed 100%; token counts are still zero. The author is the developer behind greatVibe and offers a free sandboxed guest workspace to try it.
More from coding & agent
- Claude cut a 2.6-hour video export to 20 minutes — but only after being told to rebuild the server — byhewar · 2026-10-23
- 3 flight-booking rules my AI agent checks that I'd never do by hand — Efistoffeles · 2026-10-23
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23
- How do you audit AI agent decisions in regulated workflows? A field survey — Ok_Way_1993 · 2026-10-23