运维测试
- 作者仓库星标 0
- 作者更新于 实时读取
- 作者仓库 skills-registry
- 领域
- 运维部署
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @tomevault-io · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 需手动接入
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- Docker
- 底层运行要求
- Python · Docker
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- 网络行为
- 仅限本地
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生,可能与作者实际意图存在差异。
需要注意: 未限定 allowed-tools,默认拥有全部工具权限。
---
name: copilotkit-demo-parity
description: Keeps examples/integrations/* demos aligned to the north-star (langgraph-python). Use when the u…
category: 运维部署
runtime: Python / Docker
---
# copilotkit-demo-parity 输出预览
## PART A: 任务判断
- 适用问题:部署、CI、环境检查、发布或运维排障。
- 输入要求:目标材料、限制条件、期望输出和验收方式。
- 证据边界:围绕“When this skill fires / Ground truth / Procedure: sync from north-star to instance(s)”读取原文规则,不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断:确认你的需求是否属于部署、CI、环境检查、发布或运维排障,并标出输入、限制和预期结果。
- **02** 执行计划:优先按“When this skill fires / Ground truth / Procedure: sync from north-star to instance(s)”拆成步骤,说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果:给出可复制的命令、文件改动、检查清单或内容草稿,并说明如何继续迭代。
- **04** 风险边界:结合 读取文件、写入/修改文件、主要在本地完成、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件;主要在本地完成;通常不需要额外 API Key。
- 先小样例验证,再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。 原文没有稳定的斜杠命令要求。安装验证后通常全局生效,直接在对话里点名这个 Skill 并描述任务即可。
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是:读取文件、写入/修改文件。
先用一个小任务确认它会围绕“When this skill fires / Ground truth / Procedure: sync from north-star to instance(s)”工作;涉及文件或命令时,先看 diff、日志、预览或测试结果。
检查最终产物是否包含明确结果、必要证据和下一步动作;如果输出泛泛而谈,就补充输入、边界和验收标准后重跑。
---
name: copilotkit-demo-parity
description: Keeps examples/integrations/* demos aligned to the north-star (langgraph-python). Use when the u…
category: 运维部署
source: tomevault-io/skills-registry
---
# copilotkit-demo-parity
## 什么时候使用
- 把部署运维方向的常用动作沉淀成 Agent 可调用的技能 适合处理部署、CI、发布、回滚、环境检查和运维排障,核心价值是把输入、判断、执行、验证和交付边界固定下来,避免 Agent 泛泛回答。 把任务拆成可执行、可检查、可继续迭代的步骤…
- 面向部署、CI、环境检查、发布或运维排障,优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「When this skill fires / Ground truth / Procedure: sync from north-star to instance(s)」组织步骤,不把推断写成作者事实。
- 读取文件、写入/修改文件;主要在本地完成;通常不需要额外 API Key。
- 先跑小样例,确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown,不编造命令、平台或依赖。 作者原文负责流程事实;仓库文件负责来源和命令;流狐只补充适用场景、限制和质量判断。
skill "copilotkit-demo-parity" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> When this skill fires / Ground truth / Procedure: sync from north-star to instance(s)
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> Python / Docker | 读取文件、写入/修改文件 | 主要在本地完成
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
} CopilotKit demo parity
Keeps the three (soon more) integration demos under examples/integrations/
aligned. langgraph-python is north-star; every other entry is an instance
that tracks it via examples/integrations/_parity/manifest.json.
When this skill fires
- User says "sync demos", "align integrations", "port to north-star", "parity check", "run parity sync".
- User edits a file under
examples/integrations/langgraph-python/**that is listed inmanifest.json→ tracked.verbatimFiles. - User is about to add a new integration under
examples/integrations/. pnpm parity:verifyis failing locally or in CI.
Ground truth
Read these first before making any parity decision — they override anything memorized:
examples/integrations/_parity/manifest.json— what is tracked, what is allowed to diverge, per-instance overrides.examples/integrations/_parity/canonical/PROMPT.md— the canonical system prompt. Every agent inlines this as a string literal in source. The verifier greps the canonical first line against each agent's source.examples/integrations/_parity/README.md— human-facing how-to.
Procedure: sync from north-star to instance(s)
Use when north-star changed and instances need to catch up, OR when a new instance is being bootstrapped.
- Confirm the north-star change is intentional. If the user changed
a file under
langgraph-python/, ask whether it should propagate. Changes toagent/are never auto-propagated. - Dry-run first.
Read the report: file count, rewritten package.json keys, prompt update.pnpm parity:sync --target=<instance> --dry-run - Apply.
pnpm parity:sync --target=<instance> # or --all to sync every instance - Resolve manual-merge zones. Sync does NOT touch:
agent/**— port agent code by hand. Seetracked.agentSurface.toolNamesandstateKeysinmanifest.jsonfor what must be present.src/app/api/copilotkit/**— api route differs across instances (north-star=LangGraphAgent, Docker instances=HttpAgent).Dockerfile,docker/Dockerfile.agent,serve.py,scripts/**— language-specific. Each instance keeps its own.
- Verify.
Exit 0 = green. Exit 1 = unresolved drift; fix until green.pnpm parity:verify --target=<instance>
Procedure: verify (CI-equivalent)
pnpm parity:verify # all instances
pnpm parity:verify --target=<instance>
pnpm parity:verify --json # machine-readable
Output kinds:
verbatim-file— byte-compare against north-star.package-json— tracked key mismatch vs expected value.prompt— canonical prompt first line not found in instance agent source. Instance agent needs the prompt inlined as a string literal.agent-tool— tool name from manifest not found in instance agent source. Usually means a tool was renamed or dropped; update manifest OR port the tool.agent-state— state key from manifest not found (warn only).
Procedure: add a new instance
- Create the demo dir under
examples/integrations/<name>/with the usual Next.js frontend root and anagent/subdir. - Add an entry to
manifest.json→instances:"<name>": { "role": "instance", "agent": { "language": "...", "runtime": "..." }, "allowedDivergence": [ "agent/**", "src/app/api/copilotkit/**", "Dockerfile", "docker/Dockerfile.agent", "serve.py", "scripts/**" ], "packageJsonOverrides": { "scripts.dev:agent": "<invocation>", "scripts.install:agent": "<install>" } } pnpm parity:sync --target=<name>- Port agent code (tool implementations, state schema, prompt loading).
Agent must read prompt from
agent/PROMPT.mdat startup. pnpm parity:verify --target=<name>until green.
Procedure: change the canonical prompt
- Edit
examples/integrations/_parity/canonical/PROMPT.md. pnpm parity:sync --all— writes the new prompt to every instance'sagent/PROMPT.md.- If the prompt change requires new tools or state keys, update
manifest.json→tracked.agentSurfaceto match, then port each instance's agent.
Procedure: change the tracked surface
To track a new file or key:
- Edit
manifest.json→tracked.verbatimFilesortracked.packageJsonPaths. - Run
pnpm parity:verify— instances will flag as drifted until synced. pnpm parity:sync --allto apply.
To stop tracking something, remove it from manifest.json. The verifier
and sync stop touching it. Any drift that already exists stays.
Red flags
| Signal | What it means | Do instead |
|---|---|---|
| "Just copy the file manually" | Bypasses the manifest | Add to tracked.verbatimFiles then sync |
| "Add a try/catch so verify doesn't fail" | Silencing drift | Resolve the drift or declare divergence |
"Move this into allowedDivergence to unblock" |
Scope creep | Only add to divergence with explicit reason |
pnpm parity:sync on north-star directly |
Overwrites canonical | Sync is instance-only; script refuses |
| Agent code copied between languages | Doesn't work | Port by hand; tracked.agentSurface names what |
Related skills
copilotkit-integrations— framework-specific wiring (LangGraph, CrewAI, Mastra, etc.) when bootstrapping a new instance.copilotkit-dev-workflow— monorepo conventions, Nx, commit format.docker-ci-safety— Dockerfile safety when editing per-instance container builds.
Source: CopilotKit/CopilotKit — distributed by TomeVault.
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核