Model Council Pro
v1.0.0Multi-model consensus system — send a query to 3+ different LLMs via OpenRouter simultaneously, then a judge model evaluates all responses and produces a win...
⭐ 0· 634·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description claim: send queries to multiple LLMs via OpenRouter and have a judge model pick/synthesize results. Declared requirement: OPENROUTER_API_KEY. The Python script calls openrouter.ai endpoints and implements parallel calls + a judge step. All required resources align with the stated purpose.
Instruction Scope
SKILL.md and the script only instruct the agent to call OpenRouter endpoints, aggregate responses, and have a judge model evaluate them. The code reads only the declared OPENROUTER_API_KEY env var, performs network calls to openrouter.ai, and prints or emits JSON results. It does not read arbitrary local files, other env vars, or reach out to unexpected third-party endpoints.
Install Mechanism
No install spec; this is an instruction-only skill with an included Python script. No external archives/downloads or package installs are performed by the skill itself, minimizing install-time risk.
Credentials
Only OPENROUTER_API_KEY is required and used. This is proportional: the key is needed to route queries to multiple models and will be used in Authorization headers. Note: the key grants the skill the ability to make API calls (and incur costs) on the user's OpenRouter account and may expose prompts/responses to OpenRouter's logs—this is expected for the stated purpose but is an important operational/privacy consideration.
Persistence & Privilege
always:false and default invocation settings are used. The skill does not persist or modify other skills or system configuration; it only performs runtime network calls. No elevated or persistent privileges are requested.
Assessment
This skill appears to do what it says: it uses your OPENROUTER_API_KEY to call OpenRouter and aggregate multiple model outputs. Before installing, consider: (1) Cost and billing — calls to multiple models (and a judge model) can incur non-trivial charges; monitor usage and set quotas. (2) Data exposure — prompts and model outputs are sent to OpenRouter and the upstream model providers and may be logged; avoid sending secrets or sensitive data. (3) Verify the OPENROUTER_API_KEY you provide has appropriate scope/limits and rotate keys if you later remove the skill. (4) If you need stricter privacy, confirm OpenRouter's logging/retention policy or run a trusted in-house routing endpoint. Overall the skill is coherent and consistent with its purpose; proceed if you accept the cost and data-exposure tradeoffs.Like a lobster shell, security has layers — review code before you run it.
agxntsixvk975xcrns7ygd1b65j6sgcww91817435latestvk975xcrns7ygd1b65j6sgcww91817435
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🏛️ Clawdis
EnvOPENROUTER_API_KEY
Primary envOPENROUTER_API_KEY