Prompts As Playbooks How Infrastructure Teams Codify Operational Knowledge
Source
cline.bot
Author
Cline evaluation and engineering blog
Date
Why it matters
Converting runbooks into tested, version-controlled prompts makes each step executable by an 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 → and usable by an engineer who did not write it, instead of documentation that cannot enforce anything.
Key takeaways · AI-distilled
Gardner says .clinerules markdown files in a project's .clinerules/ directory are appended to Cline's system promptThe standing instructions a model receives before any user input — defining its role, rules, tools, and tone for the whole conversation.Full definition → when toggled on, so team review criteria such as Terraform naming, provider pinning and IAM rules go through the same PR review as code.
Cline workflows live in .clinerules/workflows/ and run only when invoked as /name.md; they mix prose with XML tool tags, where execute_command runs a command verbatim and ask_followup_question pauses for an explicit approval choice.
For enforcement beyond instructions, Cline Hooks are scripts that run at lifecycle points and return JSON to allow or cancel; the article's example is a PreToolUse agent hookA script the harness runs automatically at a fixed point in an agent's loop — before a tool call, after an edit, on session start.Full definition → blocking kubectl apply unless kubectl diff ran first.
The sample triage playbook orders diagnosis before remediation: scope blast radius, build a timeline, find the change, then propose fixes stating what changes, the rollback path and how success is verified, with a dry run before any apply.
Gardner suggests testing a first playbook on a real recurring task: Cline asking clarifying questions shows where it is too vague, and Cline following it literally when the situation differs shows where it is too rigid.
Terms in this piece · Glossary
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.
system prompt — The standing instructions a model receives before any user input — defining its role, rules, tools, and tone for the whole conversation.
agent hook — A script the harness runs automatically at a fixed point in an agent's loop — before a tool call, after an edit, on session start.
