CacheForge Setup for Openclaw

A streamlined onboarding tool to register, configure, and integrate CacheForge as an LLM proxy for AI agents.

tkuehnl
v1.0.0
Feb 19, 2026
0
0
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install cacheforge-setup

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install cacheforge-setup using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is CacheForge Setup?

CacheForge Setup is a specialized utility designed to simplify the initial configuration of the CacheForge proxy environment. It allows developers to register accounts, link upstream providers like OpenAI or Anthropic, and generate API keys in under 30 seconds. By using this tool within the Openclaw Skills ecosystem, users can achieve zero-code-change integration for their AI agents, ensuring token optimization and better cost management through a single centralized gateway.

The skill eliminates the friction of manual configuration by auto-detecting existing environment variables and providing a one-line setup for OpenAI-compatible tools. Whether you are using Claude Code, Cursor, or custom agent frameworks, this skill ensures your traffic is correctly routed through the CacheForge infrastructure for enhanced performance and observability.

CacheForge Setup Use Cases

  • Initializing a new CacheForge account and obtaining a unique tenant API key.
  • Connecting existing LLM provider keys from OpenRouter, Anthropic, or OpenAI to a central proxy.
  • Automatically generating and applying configuration snippets to local settings for Openclaw Skills.
  • Validating proxy connectivity and testing model routing before production deployment.

How CacheForge Setup Works

  1. Environment scanning identifies existing API keys from providers like Anthropic or OpenRouter to infer the required upstream configuration.
  2. User credentials or invite codes are collected to register a new account or authenticate an existing session.
  3. The skill provisions a new CacheForge API key by communicating with the provisioning API endpoint.
  4. A validation step runs a test request through the proxy to ensure the upstream provider is responding correctly.
  5. The skill generates an integration snippet specifically for Openclaw Skills to automate the tool's local setup and model registration.

CacheForge Setup Setup

To begin the setup, ensure you have python3 installed on your system. Run the provisioning command to register your account and link your provider:

python3 skills/cacheforge-setup/setup.py provision \
  --email [email protected] \
  --password "your_password" \
  --upstream-kind custom \
  --upstream-key fw_...

Once provisioned, you can apply the configuration directly to your Openclaw Skills environment:

python3 skills/cacheforge-setup/setup.py openclaw-apply --set-default

CacheForge Setup Data Schema & Taxonomy

The skill manages configuration via environment variables and local JSON files to ensure seamless integration with Openclaw Skills.

Variable Description
CACHEFORGE_API_KEY The generated tenant key used for proxy authentication (prefixed with cf_).
CACHEFORGE_BASE_URL The API endpoint for the CacheForge service (defaults to https://app.anvil-ai.io).
CACHEFORGE_INVITE_CODE Optional code required for registration on invite-only deployments.
UPSTREAM_BASE_URL The base URL for the LLM provider being proxied.

Local configuration is typically stored or updated in ~/.openclaw/openclaw.json for persistence across agent sessions.

CacheForge Setup Advanced Features

  • Intelligent auto-detection of upstream providers based on API key prefixes (e.g., sk-ant- for Anthropic).
  • Vault Mode support which virtualizes tool outputs to save tokens when fetch-capable tools are detected.
  • Automated model registration for OpenRouter users, enabling instant switching between popular models.
  • Integrated validation tools to confirm the proxy and upstream provider handshake is successful.

SKILL.md


Loading

METADATA

Requires
Bins python3
Github Stars: 0
forks: 0

Featured*