How Many Credentials Should Your AI Agent Have? Zero. — Jim Clark, Docker
Source
youtube.com
Author
AI Engineer
Date
Why it matters
Gives a pattern for limiting what agents can touch: split workflows into sandboxes that never mix untrusted input with dangerous tools, and route tools through one gateway. Credentials are issued only when needed.
Key takeaways · AI-distilled
Clark argues that an agent harnessThe scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.Full definition → is a simple loop, so safety comes from controlling which context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → and tools flow into it. That makes the sandboxAn isolated environment where AI-generated code or agent actions run without being able to touch anything real.Full definition →, not the harness, the boundary to manage.
His newsroom example splits one workflow into separate researcher, fact-checker and publisher sandboxes, and his coding-AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → example grants signing keys only while the agent is committing.
Giving each sandbox a single MCP gateway endpoint creates one control point for tools, resources and prompts, and Clark says it also makes harnesses interchangeable.
Cross App Access (XAA) with Okta ties agent identity and authorization grants to an organization's existing SSO, which supports Clark's position that the right number of credentials stored in a sandbox is zero.
Terms in this piece · Glossary
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
sandbox — An isolated environment where AI-generated code or agent actions run without being able to touch anything real.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
videoCooking with Codex — Charlie Guo & Gabriel Chua, OpenAI
videoHow AI Agents Pay: Checkout in ChatGPT and Google AI Mode — Sam Parsons, PayPal
videoAdvanced Workshop: Mastering AI Observability — Doug Guthrie, Braintrust
videoFrom Vibes to Production: Evaluating and Shipping AI Agents That Work 101 — Laurie Voss, Arize AI
videoYOLO Mode, Safely: MicroVM Sandboxes for Any Agent — Rowan Christmas, DockerAI Engineer
videoUnlock Agent Autonomy: The Runtime for AI-Native Systems — Tushar Jain, DockerAI Engineer
videoAgents Without Code: Skills, YAML, and Filesystems Replaced Python — Philipp Schmid, Google DeepMindAI Engineer