Smart Router for Ollama
v1.0.0Intelligent task routing between local and cloud Ollama LLM instances. Use when the user wants cost-efficient AI responses by routing simple tasks to a local...
⭐ 0· 18·0 current·0 all-time
bySimon@simoncatbot
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (smart routing between local/cloud Ollama) matches the included Python modules and SKILL.md. The code implements system profiling, classification, health checks, routing, caching, and execution against Ollama API endpoints. No unrelated credentials or binaries are requested.
Instruction Scope
SKILL.md instructs running the included scripts (system_profiler.py, health_check.py, route.py). The runtime instructions and code operate within that scope. Note: the skill persistently logs routing decisions and stores conversation history and classification caches in local SQLite files (cache/router.db, cache/conversations.db, logs/router.log); these will contain user prompts and metadata and should be considered persistent sensitive data.
Install Mechanism
There is no install spec (instruction-only install). A requirements.txt lists only pyyaml and requests, which is proportionate to reading YAML and performing HTTP requests. No remote archive downloads or unusual installers are present.
Credentials
The skill requests no environment variables or credentials. It does make network calls to configured Ollama base_url(s) (local or user-provided cloud URLs). Because the cloud/base_url values come from config/router.yaml, you should ensure any remote endpoint is trusted; the package itself does not include unexpected external service tokens.
Persistence & Privilege
The skill does not request elevated platform privileges or always-on behavior. It writes logs and SQLite caches to the agent's working directory, creating persistent records of prompts, classifications, and endpoint health. This is normal for a router but has privacy implications.
Assessment
This skill appears to do what it claims, but take these precautions before installing or using it: 1) Inspect and update config/router.yaml so any cloud base_url points to servers you control or trust (replace placeholder addresses). 2) Be aware that prompts, conversation history, classification results, model availability, and health checks are stored locally (cache/*.db, logs/router.log); disable or clear caching/logging if you don’t want persistent records. 3) Run the code in a controlled environment or sandbox if you will route sensitive data to remote endpoints. 4) If you don't want automatic web-search integration, ensure search integration is disabled or absent. 5) Review network policies if you must prevent the skill from contacting external hosts.Like a lobster shell, security has layers — review code before you run it.
latestvk97d4haydfs9cs722rcsapkw4184qtsy
License
MIT-0
Free to use, modify, and redistribute. No attribution required.