← Back to Skills Marketplace
Openclaw Config Master
by
sniper-one
· GitHub ↗
· v1.3.3
· MIT-0
241
Downloads
0
Stars
1
Active Installs
8
Versions
Install in OpenClaw
/install openclaw-config-master
Description
Edit and validate OpenClaw Gateway config (openclaw.json / JSON5). Covers all config areas — gateway, agents, channels, models, auth, tools, commands, sessio...
Usage Guidance
Plain-language checklist for installing/evaluating this skill:
- What it actually does: it helps edit/validate OpenClaw config, offers schema lookup, can run openclaw/gateway CLI commands, create backups, restore and restart the Gateway. That's consistent with the name.
- Metadata mismatch: the registry lists no required binaries or config paths, but the skill clearly expects the openclaw/gateway CLI and access to your OpenClaw config (~/.openclaw/openclaw.json) and logs. Treat those as required even if not declared.
- Review scripts before use: the package includes shell scripts (backup/validate/restore). Inspect their contents for any destructive or network behavior before executing them.
- Protect secrets: the docs show examples referencing API keys and AppSecrets. Do not paste secrets into configs without using secure secret providers; prefer env/file-backed secret providers and check who can read those files.
- Prefer manual confirmation and a test environment: avoid allowing autonomous invocation to apply changes automatically in production. If possible run this skill in a sandbox/test instance first and make a full config backup before any change.
- Limit scope: restrict the skill's ability to run commands or restart services except when you explicitly approve a change. Ensure 'config.patch' flows require interactive consent.
If you want, I can (1) list the shell scripts' contents and point out risky commands, or (2) produce a short procedure to safely validate and run this skill in a test environment.
Capability Analysis
Type: OpenClaw Skill
Name: openclaw-config-master
Version: 1.3.3
This skill bundle provides a robust set of administrative tools for managing OpenClaw Gateway configurations. It includes well-structured bash scripts for backing up, restoring, and validating the 'openclaw.json' configuration file, alongside extensive documentation covering schema definitions, migration paths, and common error patterns. The instructions in SKILL.md promote a safety-first workflow, emphasizing schema validation and official documentation checks before applying changes. No malicious patterns, data exfiltration attempts, or harmful prompt injections were identified; the skill's capabilities are entirely consistent with its stated purpose of configuration management.
Capability Tags
Capability Assessment
Purpose & Capability
The name/description match what the files implement: schema-first editing, validation, backup/restore and migration helpers for OpenClaw config. However the skill metadata declares no required binaries, env vars, or config paths while the SKILL.md and included scripts clearly expect the openclaw/gateway CLI and access to local config files (e.g., ~/.openclaw/openclaw.json) and runtime logs. That mismatch (omitted prerequisites) is an incoherence to be aware of.
Instruction Scope
SKILL.md instructs the agent to run CLI commands (gateway config.patch, openclaw validate/doctor, openclaw gateway restart), to read system logs (tail /tmp/openclaw/openclaw-*.log) and to edit config files directly for protected paths. Those actions are within the stated goal (config management) but are high-privilege operations: they read/write sensitive configuration and secrets and can restart services. The instructions do not always require explicit human confirmation (some flows auto-apply/restart), which increases risk.
Install Mechanism
No install spec (instruction-only + helper scripts). No remote downloads or package installs are declared. This is lower risk than arbitrary install steps, but the included shell scripts will be written to disk when the skill is installed and could be executed by an agent.
Credentials
The skill does not declare any required environment variables in registry metadata, yet the docs and examples routinely reference env-based API keys, .env files (~/.openclaw/.env), and file-backed secrets for channels/providers. Requesting API keys for channel/model providers is expected by purpose, but the metadata omission (no declared env/config paths) means you cannot rely on the registry info to understand what secrets the skill will touch.
Persistence & Privilege
always: false (good) and autonomous invocation is allowed (default). Because the skill contains scripts that back up, validate, restore, and can direct agents to run commands that write config and restart the Gateway, autonomous invocation increases blast radius. This is not inherently malicious for a config-management skill, but combined with the metadata omissions and auto-apply semantics it warrants caution (prefer user confirmation and test environment).
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install openclaw-config-master - After installation, invoke the skill by name or use
/openclaw-config-master - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.3.3
补充受保护路径说明
v1.3.2
合并本地增强:Telegram/飞书详细配置模板(安全模式+开放模式)、AppSecret 安全存储、channels-config.md 新增
v1.3.1
移除真实用户ID,使用占位符
v1.3.0
完全体重写:覆盖全部17个顶层配置key,在线config.patch修改模式,官方文档必查流程,tools/session/hooks/secrets/acp/messages完整参考
v1.2.0
补全 channels 配置文档(telegram/feishu/discord/slack)+ agents.list + 安全字段(models.providers/auth.profiles/session/approvals 等),适配度 65% → 90%
v1.1.0
Update for OpenClaw 2026.3.28: add 40+ missing agents.defaults fields, restructure tools.web.search, add acp/mcp/approvals sections
v1.0.1
- Simplified and condensed documentation for easier reading, especially in Chinese.
- Added a new summary section ("一句话") and core points ("核心要点") for quick understanding.
- Grouped workflows, guardrails, $include, and upgrade process into clearer, concise sections.
- Reduced redundancy and moved some details to task-oriented, example-driven sections.
- Preserved all key information with a focus on schema validation and safe editing practices.
v1.0.0
- Initial release of **openclaw-config-master** skill.
- Provides safe, schema-first editing and strict validation for OpenClaw Gateway config (`openclaw.json` / JSON5).
- Includes detailed workflow for identifying the active config, fetching authoritative schemas, safe minimal edits, and robust validation practices.
- Documents complex config operations, guardrails for schema errors, and strategies for modular configs using `$include`.
- Offers upgrade/migration checklists, troubleshooting processes, and links to deep-dive guides and scripts for advanced use.
Metadata
Frequently Asked Questions
What is Openclaw Config Master?
Edit and validate OpenClaw Gateway config (openclaw.json / JSON5). Covers all config areas — gateway, agents, channels, models, auth, tools, commands, sessio... It is an AI Agent Skill for Claude Code / OpenClaw, with 241 downloads so far.
How do I install Openclaw Config Master?
Run "/install openclaw-config-master" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Openclaw Config Master free?
Yes, Openclaw Config Master is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Openclaw Config Master support?
Openclaw Config Master is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Openclaw Config Master?
It is built and maintained by sniper-one (@aqbjqtd); the current version is v1.3.3.
More Skills