← 返回 Skills 市场
hith3sh

RunPod

作者 Hithesh Jay · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ✓ 安全检测通过
54
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install runpod-compute
功能描述
Inspect compute resources, pods, templates, clusters, and serverless infrastructure in RunPod - powered by ClawLink.
使用说明 (SKILL.md)

RunPod

Work with RunPod from chat - inspect compute resources, pods, templates, clusters, endpoints, and account settings.

Powered by ClawLink, an integration hub for OpenClaw that handles hosted connection flows and credentials so you don't need to configure RunPod API access yourself.

Quick start

  1. Install the verified ClawLink plugin: openclaw plugins install clawhub:clawlink-plugin
  2. Start a fresh OpenClaw chat if the plugin was just installed and ClawLink tools are not visible yet
  3. If ClawLink is not configured, call clawlink_begin_pairing
  4. Tell the user to open the returned pairing URL, sign in to ClawLink if needed, and approve the device
  5. After the user confirms approval, call clawlink_get_pairing_status
  6. Tell the user to connect RunPod at claw-link.dev/dashboard?add=runpod
  7. When the user confirms RunPod is connected, call clawlink_list_integrations and then clawlink_list_tools with the runpod integration slug

Setup details

Installing the plugin

If the ClawLink plugin is not installed yet, tell the user to run:

openclaw plugins install clawhub:clawlink-plugin

If the current chat started before the plugin was installed and ClawLink tools are still unavailable, tell the user to start a fresh chat so OpenClaw reloads the plugin tool catalog.

Pairing ClawLink

If ClawLink reports that the plugin is not configured, the plugin has not been paired with the user's ClawLink account yet.

  1. Call clawlink_begin_pairing.
  2. Tell the user to open the returned pairing URL in their browser.
  3. The user signs in to ClawLink if needed and approves the OpenClaw device.
  4. After the user confirms approval, call clawlink_get_pairing_status to finish local setup.

The resulting device credential is stored locally in OpenClaw's plugin config and is only sent to claw-link.dev. The user should not paste raw credentials into chat.

Connecting RunPod

Tell the user to open https://claw-link.dev/dashboard?add=runpod and connect RunPod there. The page opens the add-connection panel filtered to RunPod. ClawLink's hosted page runs the RunPod provider connection flow. When they confirm it is done, call clawlink_list_integrations to verify, then call clawlink_list_tools with integration runpod.

Using RunPod tools

ClawLink provides tools dynamically based on what the user has connected. You do not need to know tool names or schemas in advance.

Discovery

  1. Call clawlink_list_integrations to confirm RunPod is connected.
  2. Call clawlink_list_tools with integration runpod.
  3. Treat the returned list as the source of truth. Do not guess or assume what tools exist.
  4. If the user describes a capability but the exact tool is unclear, call clawlink_search_tools with a short query and integration runpod.
  5. If no RunPod tools appear, direct the user to https://claw-link.dev/dashboard?add=runpod.

Execution

  1. Call clawlink_describe_tool before using an unfamiliar tool, before any write, or when the request is ambiguous.
  2. Use the returned schema, whenToUse, askBefore, safeDefaults, examples, and followups.
  3. Prefer account, GPU, CPU, pod, and configuration reads before creating infrastructure or secrets.
  4. For writes or anything marked as requiring confirmation, call clawlink_preview_tool first, then confirm with the user.
  5. Execute with clawlink_call_tool.
  6. If it fails, report the real error. Do not invent results or restate the failure as a missing capability unless the live catalog supports that conclusion.

What you can do

Typical RunPod tasks (actual availability depends on the user's connected account, permissions, scopes, and current ClawLink tool catalog):

  • Inspect account details, GPU types, CPU types, and pods
  • Review cluster-compatible compute options
  • Create clusters, templates, or serverless endpoints after confirmation
  • Manage secrets and private registry authentication after confirmation
  • Update user settings after confirmation

Rules

  • Always use ClawLink tools for RunPod. Do not ask the user for separate RunPod credentials.
  • Do not claim a capability is missing without checking the live ClawLink catalog in the current turn.
  • Do not invent slash commands or ask the user to paste raw credentials.
  • Ask for confirmation before creating infrastructure, storing secrets, changing registry auth, or updating settings.
  • If RunPod is not connected, direct the user to https://claw-link.dev/dashboard?add=runpod.
  • Never echo or repeat the user's ClawLink credential.

Resources

安全使用建议
Use this skill only if you trust ClawLink to broker access to your RunPod account. Review every preview carefully before approving actions that create paid infrastructure, change account settings, or store secrets.
能力评估
Purpose & Capability
The skill's RunPod account, compute, pod, template, cluster, endpoint, secret, registry, and settings operations match its stated purpose, though some supported actions can affect paid cloud infrastructure.
Instruction Scope
The instructions tell the agent to discover live ClawLink tools, describe unfamiliar tools, preview writes, and ask for confirmation before creating infrastructure, storing secrets, changing registry auth, or updating settings.
Install Mechanism
Installation requires the separate ClawLink plugin and a hosted ClawLink pairing flow; this dependency is clearly disclosed, and the artifact contains only one non-executable SKILL.md file.
Credentials
RunPod account access and cloud-resource mutation are sensitive but proportionate to the skill's purpose and constrained by discovery, preview, and confirmation guidance.
Persistence & Privilege
The artifact says a ClawLink device credential is stored locally in plugin config and sent only to ClawLink; no hidden background worker, bundled executable, or raw credential-pasting flow was found.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install runpod-compute
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /runpod-compute 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v0.1.0
Initial release of runpod-compute skill. - Enables inspection of compute resources, pods, templates, clusters, and serverless infrastructure in RunPod via ClawLink integration. - Outlines step-by-step setup, pairing, and connection process with ClawLink—no direct RunPod credential entry required. - Provides operational guidance covering discovery, execution, and confirmation rules for all RunPod tools accessed through ClawLink. - Includes best practices for safe operations, avoiding raw credential handling, and ensuring up-to-date tool catalog usage. - Lists helpful resources for both ClawLink and RunPod documentation.
元数据
Slug runpod-compute
版本 0.1.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

RunPod 是什么?

Inspect compute resources, pods, templates, clusters, and serverless infrastructure in RunPod - powered by ClawLink. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 54 次。

如何安装 RunPod?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install runpod-compute」即可一键安装,无需额外配置。

RunPod 是免费的吗?

是的,RunPod 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

RunPod 支持哪些平台?

RunPod 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 RunPod?

由 Hithesh Jay(@hith3sh)开发并维护,当前版本 v0.1.0。

💬 留言讨论