Routing Optimization
v1.0.0基于调度日志数据收集、分析和A/B测试，动态优化routing规则以提升调度命中率和准确性。
⭐ 0· 71·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (routing optimization via log collection, A/B testing, and priority adjustment) matches the SKILL.md and config.json. There are no unrelated required env vars, binaries, or install steps.
Instruction Scope
SKILL.md contains analysis and optimization functions (hit-rate, accuracy, A/B testing, dynamic priority) and PowerShell examples. It does not instruct the agent to read arbitrary system files, environment secrets, or send data to external endpoints. However, the document is example-driven and somewhat abstract: it assumes the availability of dispatch_logs and helper functions (match_route, get_base_priority, calculate_avg_accuracy, etc.) without specifying how logs are obtained or where changes are applied. That vagueness means the integration point (where logs come from and how rules are written back) should be reviewed before deployment.
Install Mechanism
This is an instruction-only skill with no install spec and no code files executed by the platform, so nothing is written to disk by an installer and there are no third-party downloads or package installs to review.
Credentials
The skill declares no required environment variables, credentials, or config paths. Its needs (access to dispatch logs and the ability to apply routing changes) are proportional to the stated purpose, but granting those accesses is the primary operational risk and should be scoped appropriately.
Persistence & Privilege
always is false and the skill does not request elevated or persistent platform privileges. The included config (config.json) sets auto-optimization to false by default, reducing the risk of autonomous changes unless the operator enables them.
Assessment
Before installing or enabling this skill, verify the following: (1) Where and how dispatch_logs will be provided (files, DB, API) and ensure the agent is granted only the minimum read access necessary; (2) Confirm and review the implementations of helper functions referenced (match_route, get_base_priority, calculate_avg_accuracy, calculate_avg_time) so you know exactly how routing rules will be evaluated and modified; (3) Keep enable_auto_optimization disabled in production until you have validated results in staging, and require human approval for any automated rule changes; (4) Check that logs do not contain sensitive PII or credentials, and if they do, sanitize or restrict access; (5) Audit any integration code that actually writes routing changes to your systems — the SKILL.md is example code and does not show how writes are performed; (6) Because this is instruction-only, there are no bundled binaries or external downloads, but you should still monitor network activity from the environment where the skill runs to detect any unexpected exfiltration. If you need higher assurance, request concrete integration code or an implementation that clearly documents data flows and access controls.Like a lobster shell, security has layers — review code before you run it.
latestvk97e1jdb1fhr8xpt2arbhccmx5840djr
License
MIT-0
Free to use, modify, and redistribute. No attribution required.