Engineer switches to Claude to simplify AI Agent building process
Understand this story
An engineer (mathaix) shared his new approach to building AI Agents: instead of relying on frameworks, orchestration engines, and multi-agent architectures, he builds a folder, uses Claude to generate AGENTS.md, holds conversations around task goals, skills, context, and a minimal toolset, and embeds the Agent into workflows via `claude -p` headless calls. He says this shift was inspired by Picasso's gradual simplification in "Bull." According to his description, headless calls mean no interactive conversation; when used for unattended tasks, tool permissions must be configured in advance; Claude can also return structured JSON for use by other programs.
AI overview based on source reports · Updated 28 min ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Hacker News AIWhat Picasso taught me about building AI Agents
The author compares learning to build AI agents to Picasso's Le Taureau, stripping away frameworks, orchestration engines and vector databases in favor of a simple folder containing an AGENTS.md, Markdown skills, context files and a minimal set of tools. Agents are invoked headless with `claude -p`, improved by reviewing session transcripts under ~/.claude/projects/<project>/<session>.jsonl, and validated with evals; subagents can be defined as Markdown files under .claude/agents/ for orchestration. The author uses this setup, with Codex and a Cloudflare MCP connection, to publish the website containing the article.
Activity for this story
Not enough continuous observations to show a trend yet.