pick-model
Skill0 starsChanged 11 days agoMITView on GitHub
Choose the right AI model for each task in a project or workflow using current prices, not memory. Finds where models are used, groups calls by task type, compares live prices from the explainx.ai MCP server, and recommends a model per task with the cost change. Use when the user asks "which model should I use", "this is too expensive", "cut my AI bill", "cheaper model for X", or sets up subagents, bots, or pipelines that call LLMs.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- jpfmscel/model-routertheir own version, first seen Oct 9, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- jpfmscel/model-router★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Search the code and config for model use: model IDs, provider SDK calls, agent and subagent definitions, cron jobs and bots.
- Distinctive: line 33
Run a small test before a big switch: the same 5 to 10 real inputs on the old and new model, side by side.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Pick model
Goal: each task runs on the cheapest model that does it well enough. The strongest model on everything wastes money. The cheapest model on everything wastes time on bad output.
1. Find the calls
Search the code and config for model use: model IDs, provider SDK calls, agent and subagent definitions, cron jobs and bots. For each call, write down:
- Task (what it does, in five words).
- Current model.
- Rough volume (per day or per run) and size (short prompt or long context).
- How bad a weak answer is: low (draft, sorted later), medium (user sees it), high (money, security, code that ships, public posts).
2. Get current prices
With the explainx.ai MCP server: call compare_model_prices for the candidate models, and search_ai_directory or get_ai_directory_entry for model details. Prices change often. Never use prices from memory. Without the server, use the providers' pricing pages and give the date you read them.
Router prices can differ from direct provider prices. Say which one you used.
3. Match model to task
| Task type | Usual fit |
| Bulk, mechanical (classify, extract, tag, reformat) | Smallest fast model that passes a 10-sample check |
| Writing people read (posts, emails, summaries) | Mid-tier model; strongest only for high-stakes text |
| Code that ships, security review, hard reasoning | Strongest model, or mid-tier plus a strong-model review |
| Long documents | The model with the best price per long-context token, not the best price per short prompt |
Run a small test before a big switch: the same 5 to 10 real inputs on the old and new model, side by side. Show the user the outputs, not only the scores.
4. Report
A table: task, current model, recommended model, monthly cost before and after (show the math), and risk. Then the total saving. Do not change config until the user agrees. When they agree, make the change in one place (a config constant or env var) so it is easy to switch back.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.