Complex Task Orchestrator
v1.2.0复杂任务编排与分治。当任务涉及批量操作（研究20+公司、处理大量数据）、多步骤工程（5+步骤有依赖关系）、sub-agent可能翻车的场景时激活。提供预记录防崩溃丢失、分治策略、超时管控、上下文膨胀防护、失败恢复方案。也适用于"任务太大不知道怎么拆"、"sub-agent老超时"、"批量操作到一半挂了"、"崩溃后...
⭐ 0· 86·0 current·0 all-time
by@xwz119
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the provided SKILL.md and helper files: the skill is an orchestrator for large/multi-step tasks and includes concrete orchestration patterns and a checkpoint utility. Nothing required (env vars, binaries) appears unrelated to its purpose.
Instruction Scope
SKILL.md instructs the agent to write logs/checkpoints and intermediate results to files (e.g., memory/YYYY-MM-DD.md, /path/to/results.json) and to prefer file-based handoff between main session and sub-agents. It also describes using Feishu APIs, Excel/openpyxl workflows, and running Python scripts for bulk processing. These operations are consistent with orchestration but mean the skill will read/write local files and may invoke executables or scripts during use — users should be aware of what data will be persisted and where.
Install Mechanism
No install spec; this is instruction-first with a small included utility script. No network downloads or installers are present in the package.
Credentials
The skill declares no required environment variables or credentials, which is reasonable. However, its instructions reference external services (Feishu APIs) and examples that will require service credentials at runtime; those credentials are not declared in the skill metadata — this is not necessarily malicious but worth noting so users provide/store external API keys securely when integrating.
Persistence & Privilege
always is false and the skill is user-invocable. The included checkpoint.py writes to a workspace under the user's home (~/.openclaw/workspace/checkpoints) — expected behavior for an orchestrator; the skill does not request elevated platform privileges or modify other skills.
Assessment
This skill appears to do what it says: orchestration patterns, checkpointing, and file-based handoff between agents. Before installing or running it, check and accept that: (1) it will create and update files in your home directory (~/.openclaw/workspace/checkpoints) and may instruct writing logs/results elsewhere (memory/YYYY-MM-DD.md or user-specified paths); review those paths so sensitive data isn't written where you don't want it; (2) if you plan to use Feishu or other external APIs mentioned, you will need to provide valid credentials separately and ensure they are stored securely (the skill does not declare or manage credentials itself); (3) the skill may suggest executing scripts (Python) for bulk work — run only if you trust the code and review the scripts (checkpoint.py is small and local-only); and (4) consider limiting the skill's access to sensitive directories and auditing created checkpoint/log files. Overall the package is internally consistent, but exercise normal caution about where data is stored and which external APIs you connect.Like a lobster shell, security has layers — review code before you run it.
latestvk975se0f5fr00ntc7cmga3wfcx847jmh
License
MIT-0
Free to use, modify, and redistribute. No attribution required.