Zuul and OpenClaw, Safe Agentic Coding Via Human Collaboration Tools
Talkat KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China 2026 · 2026-09-07..09-09·
summary
Claim: treat AI agents like untrusted contributors and use existing review and CI tooling to control them. Uses OpenClaw as agent gateway, Matrix for chat, Gerrit + OpenClaw Gerrit plugin to make each change a conversational context, and Zuul vouch pipeline so an agent must give a preliminary +1 before expensive integration CI runs. Demonstrated with WanderTracks: separate long‑lived coding and review agents iterate via Gerrit until green, agents have agent identities and keys (Matrix account, SSH/GPG), run on cloud sandboxes, and it works to cut wasted CI and human review time.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 3:30Agentic coding with guardrails
- 4:57MySQL to Drizzle and Zuul
- 7:06OpenStack growth and tooling
- 9:53Zuul’s role in CI/CD
- 13:20Speculative testing and production parity
- 15:46Zuul’s gating importance
- 16:50OpenClaw basics
- 19:41Chat-based agent interactions
- 22:08Code review automation
- 24:57Optimizing review pipelines
- 27:47Agent workflow example
- 28:30Agent identity and credentials
- 31:40Sandboxing agents
- 33:06Credential management
- 35:38Sprenbot integration
- 36:41AI agents as developers
related talks
33mGoverning the Ungovernable: Security and Compliance for AI Agents in Open Source Projec...
50mPanel: Beyond The Vibe: Code Quality First
46mVibecoding your Own Multi-Agent Workstation
30mBuilding Tools with Tools: Modern AI-Driven Development
40mAgents in Action: Advancing Open Source AI Missions
09mMulti-agent collaboration in practice: make AI code an app for you
40mHow To Tame Your Agents?
32mFrom Toil To Triumph: Harnessing Agentic AI To Streamline Infrastructure as Code