← Back to Skills Marketplace
OpenClaw Agent Compute
by
Aleksandr Krivolap
· GitHub ↗
· v0.1.7
· MIT-0
585
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install openclaw-agent-compute
Description
Public HTTP client skill exposing compute.* tools by calling a private Compute Gateway over HTTPS. Includes a starter kit to run OpenClaw preconfigured.
Usage Guidance
This skill appears to do what it says: act as an HTTP client for a private Compute Gateway. Before installing: 1) verify the MCP_COMPUTE_URL points to a gateway you control/trust (the API key will be sent there), 2) keep MCP_COMPUTE_API_KEY secret and scoped to minimal permissions on the gateway, 3) note the registry metadata omitted required env vars — ensure you set MCP_COMPUTE_URL and MCP_COMPUTE_API_KEY as shown in .env.example, 4) review the GitHub repo (README references one) and the two JS files to confirm no changes, 5) run npm install and the starter-kit in an isolated or development environment first and pin dependency versions or use audit tools if you plan production use.
Capability Analysis
Type: OpenClaw Skill
Name: openclaw-agent-compute
Version: 0.1.7
The skill provides an interface for remote command execution and file manipulation through a 'Compute Gateway' API, which are high-risk capabilities. While these functions are consistent with the stated purpose of the skill, the ability to execute arbitrary commands (computeExec) and transfer data (computeArtifactPut/Get) via scripts/client.js warrants a suspicious classification. The implementation is a transparent HTTP client using the undici library and lacks evidence of intentional malice, but the inherent risk of the provided tools requires caution.
Capability Assessment
Purpose & Capability
The name/description (compute client for a private gateway) matches the code and SKILL.md. However, registry metadata claims no required env vars while SKILL.md, .env.example, and the client code clearly require MCP_COMPUTE_URL and MCP_COMPUTE_API_KEY — this metadata mismatch should be corrected but does not indicate malicious behavior.
Instruction Scope
SKILL.md and README limit runtime actions to configuring env vars, running npm and the provided example, and using the defined compute HTTP endpoints (sessions, exec, artifacts). The instructions do not ask the agent to read unrelated files or secrets; the code only reads the two declared env vars.
Install Mechanism
There is no install spec in the registry (instruction-only), but the package includes code and a package.json with standard dependencies (dotenv, undici). Running the example requires npm install — this is low-to-moderate risk and expected for a JS client, but users should review dependencies and run in an isolated environment if concerned.
Credentials
The skill only requires a base URL and a bearer API key for the private compute gateway, which are proportionate to its purpose. Reminder: the API key grants whatever privileges the gateway exposes, so treat it as sensitive.
Persistence & Privilege
The skill does not request always: true or other elevated platform privileges. It does include a starter-kit that passes env vars into a Docker container (expected for this use case). Autonomous invocation is allowed by default but is not combined with any unusual privileges here.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install openclaw-agent-compute - After installation, invoke the skill by name or use
/openclaw-agent-compute - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.7
- Added .env.example files for both the main project and the starter kit to clarify environment variable setup.
- Included a LICENSE file.
- No changes to core functionality.
v0.1.1
- Expanded API support: added artifact management endpoints and session status checks in documentation.
- Updated usage expectations for the Compute Gateway interface.
- Added publishing instructions with new files: `PUBLISHING.md` and `CLAWDHUB_RUNBOOK.md`.
- Documentation improvements for onboarding and reference.
v0.1.0
Initial public release of openclaw-agent-compute.
- Exposes compute.* tools over HTTP by forwarding to a private Compute Gateway via HTTPS.
- Requires configuration of MCP_COMPUTE_URL and MCP_COMPUTE_API_KEY.
- Compatible with APIs for session creation, command execution, usage lookup, and session deletion.
- Includes scripts for HTTP client functionality and execution example.
- Provides a starter kit for quick setup with image override support.
Metadata
Frequently Asked Questions
What is OpenClaw Agent Compute?
Public HTTP client skill exposing compute.* tools by calling a private Compute Gateway over HTTPS. Includes a starter kit to run OpenClaw preconfigured. It is an AI Agent Skill for Claude Code / OpenClaw, with 585 downloads so far.
How do I install OpenClaw Agent Compute?
Run "/install openclaw-agent-compute" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is OpenClaw Agent Compute free?
Yes, OpenClaw Agent Compute is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does OpenClaw Agent Compute support?
OpenClaw Agent Compute is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created OpenClaw Agent Compute?
It is built and maintained by Aleksandr Krivolap (@aleksandrkrivolap); the current version is v0.1.7.
More Skills