OpenClaw Chain of Execution (CoE)

v1.0.0

OpenClaw 原生执行链(Chain of Execution, CoE)技能。让 OpenClaw 在处理复杂任务时,实时分步输出执行过程:拆解问题 → 选择 skill → 调用 skill → 调用模型 → 等待响应 → 拿到结果 → 最终汇总,完全贴合 OpenClaw 实际运行流程,让用户清楚看到每...

0· 115·0 current·0 all-time
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 claim a Chain-of-Execution tracker that emits step-by-step output. Included Python code (cot_tracker.py and a small main.py) implements exactly that behavior (printing steps, start/wait/result/done). No unrelated binaries, env vars, or network calls are requested.
Instruction Scope
SKILL.md stays within scope: it documents how to enable/disable the tracker, automatic/manual triggers, and how other skills can import and call CoETracker. The tracker prints whatever strings are passed to it (to stdout and flushed). That means any data other skills pass into tracker.step/result etc. will be echoed to the user/UI — a potential privacy consideration but not a functional incoherence.
Install Mechanism
No install spec; code is instruction-only with small local Python files. There is no remote download, package install, or extraction that would introduce supply-chain risk.
Credentials
The skill declares no required environment variables, no credentials, and the code does not access env vars or config paths. The requested scope of access is minimal and matches the stated purpose.
Persistence & Privilege
always is false and the skill does not modify other skills or system configuration. It exposes an API for other skills to import the tracker, which is appropriate for a tracing helper.
Assessment
This skill is a simple, local tracer that prints step-by-step status to stdout and requires no credentials or network access — it appears to do exactly what it says. Before enabling or letting other skills call it, be aware it will echo any text you pass into it: do not pass secrets, credentials, or sensitive data into tracker.step/result/done if you don't want that content surfaced in logs/UI. Also note that other skills must be able to import the module (filesystem/packaging), but that is a normal development consideration rather than a security problem.

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

latestvk97ezbe46qcvd7ekjtnmpvkph9839ymr

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Runtime requirements

🔗 Clawdis

Comments