#002
Minimal Agent Framework (MAF)
Every AI agent framework is a maze of abstractions. You can't trace what happened, you can't replay a failed run, and when something breaks you're debugging the framework instead of your agent. You need something you can actually read.
Inside the Vault
- One runtime loop — the entire agent core fits in your head
- Four typed tools: shell.exec, fs, http.fetch, kv — with JSON schema validation
- Full JSONL traces on every run — replay any execution deterministically
Plus step-by-step usage and direct repo access.
A practical field guide to running coding agents safely: scope, isolation, verification, and review.
A practical, execution-first guide to build, run, debug, and harden your first AI agent with tools, guardrails, and production checks.
I built MAF — a minimal AI agent framework in Python with one core loop, typed tool schemas, and JSONL traces. Here's how to build an AI agent from scratch, what broke against real APIs, and why minimal beats monolithic.
A deep, command-level teardown of claudeagentsdk (#005): an open-source agent workspace built around the Anthropic Agent SDK, with a FastAPI backend, a Vite/React frontend, and an optional Vercel Sandbox runner for async, reproducible runs.
A command-level, evidence-first teardown of where OpenClaw fits in an ai developer tools stack: architecture, workflows, incidents, throughput, and adoption boundaries.
I built datafast-cli and pointed an autonomous AI agent at it. 13 commands, 2 bugs found, and the 5 principles that make CLI tools genuinely useful as AI agent tools.