Blue Green Deployer

A specialized skill for safely managing and testing OpenClaw configuration changes using a Blue/Green deployment pattern. Use when you need to test new setti...

Security Scan

The skill's code and instructions match its stated purpose (managing OpenClaw configs via a blue/green pattern); it manipulates only the user's OpenClaw config files and invokes the OpenClaw CLI as expected.

Purpose & Capability

The name/description claim a Blue/Green deployer for OpenClaw; the provided script implements that behavior (copies, backups, creates symlinks, runs `openclaw status`, promotes or restores). Required files and actions are proportional to the purpose.

Instruction Scope

SKILL.md instructs running the included deploy.sh and using jq for validation; the script itself operates only on $HOME/.openclaw/*. It does remove and recreate the live config file (via symlink and cp) which is expected but potentially disruptive if run on a production host without testing. SKILL.md lists `jq` validation steps that are not invoked by the script (you must run them manually if desired).

Install Mechanism

Instruction-only skill with an included shell script and no install spec — nothing is downloaded or installed by the skill itself, minimizing install-time risk.

Credentials

The skill declares no required env vars or credentials and only touches $HOME/.openclaw config files. The dependency list mentions `jq` but the script does not call it; `openclaw` is used and is a reasonable dependency given the stated purpose.

Persistence & Privilege

The skill is not always-enabled and does not change other skills or global settings. It does modify files under the user's OpenClaw configuration, so the agent running the skill must be trusted to perform filesystem operations on that path. The platform default allowing autonomous invocation means the agent could run this script without explicit user action — consider invocation policy if that matters.

Assessment

This skill appears to do what it says, but it modifies your live OpenClaw config files. Before installing or running: (1) review scripts/deploy.sh yourself to confirm behavior; (2) ensure you trust the `openclaw` binary on your system (the script runs `openclaw status`); (3) take an independent backup of $HOME/.openclaw/openclaw.json in case of failures; (4) test the workflow on a non-production account or copy of your config; (5) note SKILL.md mentions using `jq` for validation but the included script does not call it — run `jq empty openclaw.json.green` manually if you want stricter JSON checks; and (6) consider restricting autonomous invocation if you don't want the agent to run the deployment script without explicit approval.

This skill provides a structured pre-production testing workflow for OpenClaw configuration changes using a Blue/Green deployment pattern.

Workflow

1. Setup the Green Environment

Initialize the testing area by creating the Green configuration from your current Blue master.

• Command: bash <skill-path>/scripts/deploy.sh

2. Experiment

Modify the files within the green configuration folder or the openclaw.json.green file directly using your preferred editor.

3. Audit and Validate

Verify that the changes in the Green environment are valid and do not break the OpenClaw service.

• Step A: Check JSON syntax: jq empty openclaw.json.green
• Step B: Run the deployment script to perform the symlink-based validation check.

4. Promote or Rollback

• To Promote: If the audit passes, the deploy.sh script will automatically promote the Green configuration to Blue.
• To Rollback: If the audit fails, the deploy.sh (or manual action) will revert the Blue configuration to its previous state using the .bak file.

Troubleshooting

• Validation Error: If the script reports a failure, check the openclaw.json.green file for syntax errors (missing commas, brackets) or invalid keys.
• Manual Check: Always use openclaw status after any deployment to ensure the live service is operational.