English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

OpenClaw: When AI Fixes Its Own Bugs, Humans Are Left with Verification

Forum topic · QianXun · 2026-06-13

Summary

This in-depth analysis explores the rise of OpenClaw, a viral AI personal agent project created by Peter Steinberger, the founder of PSPDFKit. The article examines how OpenClaw's 'files as the source of truth' architecture—centered on Markdown files like AGENTS.md and SOUL.md—acts as a 'constitution' constraining large AI models. It covers Steinberger's 'Frustration-Driven Development' philosophy, his suite of 40+ CLI tools built for AI agents rather than humans (deliberately choosing CLI over MCP), and his 'Octopool' technique for distributing API requests across edge nodes to reach 40 million tokens per minute. The piece also analyzes the Crabbox workflow, where AI autonomously reproduces, fixes, and verifies bugs in disposable sandboxes; Microsoft's MXC execution containers with declarative, auditable agent permissions; and the broader power shift from human 'makers' to human 'verifiers' as AI closes the development loop. The article frames AI slop pollution of open source, token economics (603 billion tokens, $1.3 million in 30 days), and agent sandboxing as signals of a fundamental paradigm shift in software development.

This article analyzes OpenClaw and the emerging AI-agent paradigm through the work of Peter Steinberger, translated and condensed from the original Chinese forum post.

Key Points

Who Is Peter Steinberger?

  • Steinberger founded PSPDFKit, a PDF rendering engine running on over 1 billion devices, and sold the company in 2022 for roughly €100 million before retiring.
  • In April 2025 he wanted an AI personal assistant. After loading his entire WhatsApp history into GPT-4.1's million-token context window and seeing remarkable results, he waited seven months for a lab to ship such a product. Nobody did. Frustrated, he wrote the OpenClaw prototype in one hour: message comes in → call CLI → return result to WhatsApp.
  • OpenClaw exploded on GitHub—50,000 stars within days, 315,000 stars in 84 days, surpassing React's decade-old record. After Anthropic banned premium-subscription users running ClawdBot, he renamed it OpenClaw and joined OpenAI in February 2026 to work on next-generation personal agents.
  • AI Slop and the 'Constitution' Files

  • AI-generated spam is drowning open source: curl shut down its bug bounty after finding only ~5% of AI-generated vulnerability reports were real; RubyGems saw 10x report growth with mostly invalid submissions; tldraw disabled external pull requests entirely.
  • OpenClaw's answer: "Files are the source of truth." The system is driven by 8 core Markdown files: SOUL.md (personality/ethics), AGENTS.md (the supreme constitution—workflow, collaboration, safety red lines), IDENTITY.md, USER.md, MEMORY.md, HEARTBEAT.md (autonomous check-ins every 30 minutes), TOOLS.md, and BOOTSTRAP.md.
  • Rules live in human-readable, Git-trackable text files rather than model weights. Golden rule: don't "train" your agent in chat—permanent configuration happens in Markdown files. Giving an AI an explicit constitution transforms it from a slop producer into a directed builder.
  • Frustration-Driven Development and CLI-First Tools

  • Steinberger built 40+ tools for AI agents, not humans: wacli/imsg/bird (messaging), gogcli (unified Google Workspace control), Peekaboo (screen understanding + GUI automation), Brabble (local Whisper), plus smart home, ordering, and reminder CLIs.
  • Three design principles:
  • 1. CLI over MCP — no schema loading into context, Unix pipe composition, self-correcting via --help. Per Steinberger, agents are far better at calling CLIs than MCP. 2. Zero-config auth — tools like Sweet Cookie extract browser login state so agents inherit existing permissions. 3. Agent-first interaction — JSON output by default; lenient error handling because "agents will always get parameters wrong."
  • After observing users, he concluded "80% of apps will disappear"—an AI agent with more context makes better decisions than the apps it replaces.
  • Token Laundering and AI Compounding

  • In May 2026 his personal bill: 603 billion tokens, 7.6 million requests, $1,305,088 in 30 days, mostly on GPT-5.5.
  • To exceed standard rate limits he built "Octopool": distributing large workloads across many independent edge-node API keys, each staying under per-key thresholds—effectively splitting large transactions into many small, individually compliant ones.
  • The "AI compounding" logic: each tool built multiplies time saved by exponentially growing call counts, which funds building more tools. He ships ~600 commits per day solo to OpenClaw.
  • Crabbox: AI Verifying AI

  • The Crabbox workflow: whenever investigating a bug, Codex rebuilds the exact bug state in a temporary sandbox, verifies the bug exists, fixes it, and re-verifies the fix.
  • Zero state pollution (fresh disposable environments, no "works on my machine") and massive parallelism (10+ independent sessions simultaneously).
  • Unlike traditional CI/CD, AI can see interfaces, operate UIs, and understand visual feedback. Humans only validate the final Pass/Fail report.
  • MXC: Microsoft's Execution Containers

  • At Build 2026, Microsoft announced MXC (Microsoft Execution Containers), an OS-level sandbox for AI agents with three layers: declarative YAML policies (least privilege, default deny), runtime containment (virtualized filesystems, restricted networking, hard resource caps), and full audit logging.
  • The ACS (Agent Control Specification) binds permission declarations to the agent itself, not the host environment—"the shackles follow the agent."
  • The critique: rules can never keep pace with AI evolution. If an agent learns to modify its own ACS file, who guards the last gate? The answer loops back to human verification.
  • The Power Shift: From Maker to Verifier

  • Humans no longer write code—they write constitutions (AGENTS.md).
  • AI closes the full loop: reproduce → fix → self-test, with humans approving PRs.
  • Steinberger's closing metaphors: "Programming will become like knitting—people do it because they enjoy it"; "You're not just a programmer... you're a builder." The bottleneck is no longer typing speed but the ability to precisely express intent.
> "The future belongs to programmers who can tell stories." — Peter Steinberger

Tags

#openclaw#ai-agents#peter-steinberger#cli-tools#ai-slop#microsoft-mxc#developer-tools#openai

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177981230