AGENT SKILL

oceinteg

Run named OpenClaw Enterprise end-to-end integration scenarios against real supported installations. Use only when explicitly invoked.

License 许可
MIT
最近更新
2026-11-13
Stars
★ 461
Forks
115

01 前言 Front Matter

name
oceinteg
description
Run named OpenClaw Enterprise end-to-end integration scenarios against real supported installations. Use only when explicitly invoked.
dependencies
[ "enterprise-testing" ]

02 资源文件 7

03 SKILL.md 正文

OCE integration scenarios

Use this skill only when the user explicitly invokes oceinteg <scenario> or $oceinteg <scenario>. Do not activate it for general integration-testing requests. Follow $enterprise-testing for real-runtime prerequisites, evidence, and failure classification, then read only the selected scenario below.

Invocation Scenario
oceinteg main Main acceptance test: fresh installation, both runtime presets, repository permissions, Slack, Linear approvals, and native UI continuity.

If the scenario is missing or unknown, show the available names and ask which one to run. Do not substitute another scenario or start infrastructure work. Creating or editing this skill does not run its scenarios.

Add future scenarios as references/<scenario>.md and register them here. Keep each scenario's inputs, external effects, procedure, assertions, and cleanup in its own reference. This is an agent workflow, not an installed shell command.

For main, also read Runtime and isolation acceptance and Supply credentials. These are required parts of main, not separately invokable scenarios.

Choose one setup reference per selected topology before running main:

Topology Setup
EKS with Helm OCC EKS setup
Local k3d with Helm OCC Kubernetes-only setup
Local k3d with Compose OCC Compose and Kubernetes setup

These references sequence the user-facing guides; those guides own commands and supported configuration. Setup references are not additional invocations.

04 同仓库更多 Skills