ClawHub

Cron Creator 定时任务创建器

v1.0.13

【针对飞书优化】定时任务创建指引，专门针对飞书、尤其是多 Agent 场景进行优化，解决定时任务不稳定问题。

⭐ 1· 148·1 current·1 all-time
by@jhc888007
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description, required_files (openclaw.json, current_chat_context), and required_tools (openclaw CLI, read) match the stated goal of composing/adding cron tasks for Feishu and multi-agent setups. No unrelated environment variables, binaries, or installs are requested.
✓
Instruction Scope
SKILL.md instructs the agent to read only the declared files (openclaw.json and current_chat_context) and to build/emit openclaw cron commands. It explicitly forbids fabricating missing data and restricts actions (must use openclaw cron add, not crontab or sleep). The scope stays within creating cron jobs.
✓
Install Mechanism
No install spec or external downloads — instruction-only content with no code files. This is low risk because nothing is written or executed by the skill itself beyond generating CLI commands.
ℹ
Credentials
The skill reads openclaw.json and chat context to extract account IDs and target IDs; this is necessary for Feishu multi-agent routing. However, openclaw.json may contain sensitive configuration (account identifiers or secrets) depending on local setup — the README/SKILL.md do not explicitly warn about secrets in that file. The skill requests no environment credentials, which is proportionate.
✓
Persistence & Privilege
always:false and normal model invocation; the skill does not request persistent elevated privileges or modify other skills' configs. It only prescribes how to run openclaw CLI commands.
Assessment
This skill appears coherent and focused on composing openclaw cron commands for Feishu. Before installing: (1) confirm openclaw.json does not contain sensitive secrets (API keys) or restrict access to it if it does; (2) ensure the agent is only allowed to read the declared context files and that it will ask you (not fabricate) any missing user_id/chat_id; (3) test with a non-sensitive dummy cron to verify generated commands and then inspect the created task list; (4) confirm you are comfortable the agent can run openclaw CLI commands in your environment. If openclaw.json contains secrets, avoid granting this skill access or sanitize the file first.

Like a lobster shell, security has layers — review code before you run it.

latestvk97an1a8x172rtfejd4h72zj8x842q37

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments