← Back to Skills Marketplace
ddanieli

dataclaw-setup

by ddanieli · GitHub ↗ · v0.4.0 · MIT-0
cross-platform ✓ Security Clean
79
Downloads
1
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install dataclaw-setup
Description
Setup guide for installing and configuring DataClaw itself for use with OpenClaw. Use when the user wants to learn what DataClaw is or finish initial setup....
README (SKILL.md)

DataClaw Setup

Use this skill to explain what DataClaw is and guide the user through setup for OpenClaw.

What DataClaw does

DataClaw is a localhost-only server that gives AI agents controlled access to a database through MCP.

It is designed for cases where the user wants:

  • per-agent API keys
  • approved-query workflows
  • optional raw query or raw execute permissions
  • an audit log of MCP access
  • local control over PostgreSQL or SQL Server access

Setup

Before DataClaw can be used from OpenClaw, the user must complete setup in this order:

  1. Install DataClaw from https://dataclaw.sh or https://github.com/ekaya-inc/dataclaw.
  2. Run dataclaw.
  3. Open the local DataClaw UI in the browser.
  4. Configure the datasource.
  5. Create or configure the DataClaw access point the user wants OpenClaw to use.
  6. In the DataClaw UI, use Install as a Skill for that configured access point.

The Install as a Skill action is the step that creates the real local OpenClaw integration for that access point.

Important

This skill is a setup and discovery guide only.

It does not provision the actual DataClaw MCP connection by itself.

Do not assume this skill means DataClaw is already installed, configured, or running.

The real OpenClaw-ready skill for a specific DataClaw access point is generated and installed by DataClaw after setup, for example dataclaw-marketing.

After Setup

Once the user has installed the generated DataClaw skill from the DataClaw UI:

  • use the generated access-point skill that DataClaw created
  • prefer approved queries when available
  • do not assume raw SQL access is allowed
  • do not assume write or execute permissions are allowed
  • if an operation appears unavailable, explain that the DataClaw access point may not expose that capability

If Setup Is Not Finished

If the user asks to use DataClaw before it has been installed and configured:

  • tell them to install and run DataClaw first
  • direct them to the local DataClaw UI
  • ask them to finish datasource and access-point setup there
  • tell them to use Install as a Skill in DataClaw when that option becomes available

Links

  • Website: https://dataclaw.sh
  • GitHub: https://github.com/ekaya-inc/dataclaw
Usage Guidance
This skill is a documentation-only setup guide and appears internally consistent. Before proceeding: verify you're downloading DataClaw from the official site or the linked GitHub repo; run the dataclaw server locally and confirm it binds to localhost only; when DataClaw generates the actual per-access-point skill (e.g., dataclaw-marketing), inspect that generated skill's manifest and permissions before installing it into OpenClaw (watch for environment-variable requirements or network endpoints); and review any binaries you install from dataclaw.sh or GitHub releases for expected checksums or signed releases if available.
Capability Analysis
Type: OpenClaw Skill Name: dataclaw-setup Version: 0.4.0 The skill bundle consists solely of documentation and setup instructions for DataClaw, a tool for managing database access via MCP. It contains no executable code, and its instructions in SKILL.md are purely informational, guiding the user to official resources (dataclaw.sh and GitHub) for manual installation. No malicious behaviors, data exfiltration, or prompt-injection risks were identified.
Capability Assessment
Purpose & Capability
The skill name/description and runtime instructions all describe a setup/discovery guide for DataClaw; it requires no binaries, env vars, or installs, which is appropriate for a documentation-only skill.
Instruction Scope
SKILL.md only instructs the user to install and run DataClaw, open the local UI, configure a datasource and use the UI's 'Install as a Skill' action. It does not ask the agent to read unrelated files, access secrets, or transmit data to unexpected endpoints.
Install Mechanism
There is no install specification and no code files; nothing will be written or executed by the skill itself. This is the lowest-risk delivery mechanism for a setup guide.
Credentials
The skill declares no required environment variables, credentials, or config paths and the instructions do not reference any hidden secrets—requested access is proportional to a documentation/help skill.
Persistence & Privilege
always is false and model invocation is allowed (the platform default). The skill does not request persistent presence or attempt to modify other skills or system-wide settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install dataclaw-setup
  3. After installation, invoke the skill by name or use /dataclaw-setup
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.4.0
Publish DataClaw skill version 0.4.0
Metadata
Slug dataclaw-setup
Version 0.4.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is dataclaw-setup?

Setup guide for installing and configuring DataClaw itself for use with OpenClaw. Use when the user wants to learn what DataClaw is or finish initial setup.... It is an AI Agent Skill for Claude Code / OpenClaw, with 79 downloads so far.

How do I install dataclaw-setup?

Run "/install dataclaw-setup" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is dataclaw-setup free?

Yes, dataclaw-setup is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does dataclaw-setup support?

dataclaw-setup is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created dataclaw-setup?

It is built and maintained by ddanieli (@ddanieli); the current version is v0.4.0.

💬 Comments