Step-by-step help / Edition 01
How-To Guides
Clear walkthroughs for setting up, testing, and improving agent-assisted work.
Open source / Complete sequence
Agent Setup Toolkit
Build context, access, judgment, routines, tests, and a gradual rollout plan. Includes seven editable Markdown templates.
Open the toolkit- 01Anatomy of a reliable agent promptTurn an attractive idea into an operating contract an agent can follow, test, and safely stop.
- 02Connecting tools without over-sharingVerify capability with the smallest possible permission and a harmless protected read.
- 03Human approval for consequential actionsDesign an approval moment that shows exactly what will happen and expires when the target changes.
- 04Testing and improving a bot after its first runUse a supervised first run, a small failure set, and stable acceptance criteria before scheduling.
- 05Built-in connectors, MCP, browser control, and CLI toolsChoose an execution path based on available capability, permission clarity, and evidence—not product hype.
- 06Prompt-injection and secret-handling checklistA preflight checklist for retrieved instructions, credentials, cross-system data, and safe failure.
- 07Work contracts and completion receiptsDefine the job, limit the access, and prove the result with the smallest contract the work needs.