← Back to Skills Marketplace
Openclaw Push Doctor
by
Cosmos Fang
· GitHub ↗
· v1.0.2
· MIT-0
123
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install openclaw-push-doctor
Description
openclaw-healthcheck is a self-diagnostic skill that checks and repairs openclaw's communication channels and scheduled tasks. It diagnoses Feishu/lark-cli a...
Usage Guidance
This skill is coherent for diagnosing and repairing Openclaw push channels and cron tasks, but it does read your ~/.openclaw config, logs, and system crontab and can apply crontab edits or restart services if you confirm. Before installing or running: (1) review/backup your crontab and ~/.openclaw directory; (2) don't paste tokens into the chat — follow the skill's guidance to set tokens in your terminal; (3) note the small inconsistency where scripts use ~/.openclaw rather than honoring OPENCLAW_CONFIG_DIR — adjust commands or paths if you store configs elsewhere; (4) only allow dedup/apply steps after you inspect the proposed crontab diff and explicitly confirm. If you want extra safety, run the check scripts first (diagnostic mode) and avoid the --fix/--dedup actions until you are comfortable.
Capability Analysis
Type: OpenClaw Skill
Name: openclaw-push-doctor
Version: 1.0.2
The openclaw-push-doctor skill is a diagnostic and repair utility for OpenClaw communication channels (Feishu, Telegram, WeChat) and cron tasks. The scripts (e.g., check-all.js, check-telegram.js, fix-feishu.js) function as prompt generators that guide an AI agent through system checks using standard tools like curl, pgrep, and crontab. Notably, the Telegram diagnostic script includes explicit security instructions for the agent to avoid asking the user for sensitive tokens in the chat, instead directing the user to set them locally in their environment, which demonstrates a clear lack of malicious intent regarding credential theft.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description match the scripts: checks Feishu (lark-cli), Telegram, WeChat bridge, and cron tasks. Declared binaries (curl, python3, pgrep, lark-cli, crontab) and optional env vars (TELEGRAM_BOT_TOKEN, TELEGRAM_CHAT_ID, OPENCLAW_CONFIG_DIR, FEISHU_APP_ID) are appropriate for the stated checks.
Instruction Scope
The SKILL.md and scripts instruct the agent/operator to read local files (~/.openclaw/*), inspect logs, run curl against Telegram APIs, check processes with pgrep, and (when fixing) export/edit/import system crontab lines and restart daemons. Those actions are within the tool's purpose, but they involve access to local state and potentially destructive crontab changes. Also: the SKILL.md declares OPENCLAW_CONFIG_DIR as an override, but most scripts hardcode ~/.openclaw instead of reading that env var — a minor inconsistency to be aware of.
Install Mechanism
No install spec (instruction-only) and included scripts are prompt/guidance generators (they print shell steps). There is no download-from-URL or archive extraction. Risk from install mechanism is low.
Credentials
No required secrets are demanded by the skill. Optional env vars (Telegram token/chat id, Openclaw config dir, Feishu app id) are reasonable and used in scripts. The skill explicitly warns not to paste tokens into chat and instructs the user to set tokens in their terminal — consistent with least exposure.
Persistence & Privilege
Registry metadata shows always:false and metadata.openclaw.disable-model-invocation:true (skill will not be autonomously invoked by the model), which reduces autonomous risk. The skill's runtime guidance can modify the user's crontab and local configs, but actions that change crontab are gated behind explicit user confirmation per the scripts.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install openclaw-push-doctor - After installation, invoke the skill by name or use
/openclaw-push-doctor - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
fix: remove secret-in-chat pattern (never ask user to paste token in conversation), add explicit crontab confirmation pause before applying edits
v1.0.1
fix: declare python3/pgrep as required binaries, add TELEGRAM_CHAT_ID env, set disable-model-invocation:true to reduce autonomous-invocation risk
v1.0.0
Initial release: full diagnostic for Feishu/Telegram/WeChat channels and cron tasks
Metadata
Frequently Asked Questions
What is Openclaw Push Doctor?
openclaw-healthcheck is a self-diagnostic skill that checks and repairs openclaw's communication channels and scheduled tasks. It diagnoses Feishu/lark-cli a... It is an AI Agent Skill for Claude Code / OpenClaw, with 123 downloads so far.
How do I install Openclaw Push Doctor?
Run "/install openclaw-push-doctor" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Openclaw Push Doctor free?
Yes, Openclaw Push Doctor is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Openclaw Push Doctor support?
Openclaw Push Doctor is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Openclaw Push Doctor?
It is built and maintained by Cosmos Fang (@cosmofang); the current version is v1.0.2.
More Skills