القائمة
Rules for batching write_file calls efficiently. Prevents chatty incremental file writes and ensures the artifact store is updated atomically per logical unit of work.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
ACR integration for AKS Automatic. Teaches attaching an ACR, image reference conventions (digest pinning, no :latest), and pull-secret-free authentication via the managed identity.
Writing idiomatic, safe, and reviewable Bicep templates for Azure resources.
Non-blocking cluster status peek — run at the end of Phases 3, 4, 5 to check AKS provisioning progress without hanging.
Voice, tone, and interaction patterns for Kickstart agents.
Configure Infrastructure phase playbook — launch the dedicated Kickstart cluster-setup view, which collects and creates the Azure resources (subscription, resource group, AKS Automatic cluster, ACR) and hands the results back to the chat.
Deploy phase playbook — build, push, apply with Azure CLI and kubectl.
| name | kickstart-file-generation |
| description | Rules for batching write_file calls efficiently. Prevents chatty incremental file writes and ensures the artifact store is updated atomically per logical unit of work. |
| disable-model-invocation | true |
When generating multiple related files, write them all before reporting to the user. Do not write one file, report, write another file, report — this creates noise and the user has no useful intermediate state to act on.
Compute all file contents in a single reasoning pass. Then write all files. Then report.
THINK: what files are needed and what goes in each
WRITE: file 1, file 2, file 3, … (consecutive write_file calls)
REPORT: "I generated the following files: …"
A "batch" is all files that belong to the same logical unit of work:
Do not batch files across unrelated features or stacks into a single turn — that makes validation harder.
Before writing any file:
If there are dependencies between files (e.g., the K8s deployment references an image tag that comes from the Bicep output), resolve those values before writing either file.
If a write_file call fails mid-batch:
Before overwriting an existing file in the artifact store, announce your intent:
"I'm about to overwrite
k8s/deployment.yaml— it already exists. Proceeding."
Do not silently overwrite without announcement.
Keep generated files under 100 KB each. If a single file would exceed this, split it into multiple files with clear naming (e.g., main.bicep + modules/aks.bicep).