Create a Botpress AI agent with the ADK using OpenClaw
ReviewAudited by ClawScan on May 1, 2026.
Overview
This is a coherent Botpress ADK guide, but users should notice that it asks them to install a CLI from the internet, authenticate to Botpress Cloud, and deploy bots.
Install this skill if you want guidance for Botpress ADK development. Before following its setup commands, verify the Botpress ADK installer source, log in only to the intended Botpress workspace, and review/test any bot before running deployment commands.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Running the installer can execute code on the user's machine.
The guide asks the user to install the ADK CLI by piping a remote latest-release installer into a shell. This is a common setup pattern and is central to the skill's purpose, but it means the user is trusting remote installer code.
curl -fsSL https://github.com/botpress/adk/releases/latest/download/install.sh | bash
Use the official Botpress source, inspect the installer or use a manual download/checksum when available, and run it only if you intend to install the ADK CLI.
The ADK CLI may be able to access or modify Botpress Cloud resources associated with the authenticated account or workspace.
The workflow requires authenticating with a Botpress Cloud account. This is expected for developing and deploying Botpress bots, but it grants the CLI account-level access needed for the project.
adk login # Authenticate with Botpress Cloud
Authenticate only to the intended Botpress account/workspace and avoid using unnecessarily privileged accounts for development.
A mistaken deployment could affect users interacting with the Botpress bot.
The guide includes production deployment commands. This is purpose-aligned and user-directed, but deployment can change a live bot's behavior.
adk deploy # Deploy to production when ready
Test locally first, review generated code and configuration, and deploy only when you are ready for the bot to run in production.
