agentleFS
byhowbyhow/model-picker/llms.txt

model-picker

llms.txt1 starsChanged 7 months agoApache-2.0View on GitHub

model-picker is a terminal-first model discovery tool for AI builders. Use it to compare models across price, speed, and context, pick the right model for coding agents, and access docs through GitHub, a web docs UI, and llms.txt. For full documentation, see: https://model-picker.dev/docs For human-readable README, see: ./README.md model-picker has two data modes: The pick command recommends models tailored to specific coding agents. Harness adapters MUST use pick --contract rather than the wider, command-specific --json payload. The contract envelope contains…

  • Reads credentials
  • Installs packages

We split this llms.txt into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    When --agent is set and --task is omitted, defaults to --task agent.

    line 104

≈ “Common” is an estimate: at least 1 in 5 of the 3,225 llms.txt 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
llms.txt
# model-picker

## Overview

model-picker is a terminal-first model discovery tool for AI builders. Use it to compare models across price, speed, and context, pick the right model for coding agents, and access docs through GitHub, a web docs UI, and llms.txt.

For full documentation, see: https://model-picker.dev/docs
For human-readable README, see: ./README.md

## Install

```bash
# Run instantly — no install required
npx model-picker <command>
bunx model-picker <command>

# Install globally
npm i -g model-picker
bun install -g model-picker

# After install, CLI is available as `model-picker` or `mp`
```

## Data Modes

model-picker has two data modes:

### Live (no API key required for primary path)
- top, get
- Queries OpenRouter frontend API directly, falls back to Firecrawl if available
- Mirrors OpenRouter URL query parameters

### Snapshot (offline, no API key required)
- compare, pick, export, doctor
- Uses local packaged model catalog
- Works without network access

## Commands

### Discovery

```bash
# Live OpenRouter queries
model-picker top --order most-popular --limit 10
model-picker top --categories programming --order top-weekly --zdr
model-picker top --input-modalities text,image --output-modalities image --max-price 0.5
model-picker get openai/gpt-5.4

# Local catalog queries
model-picker compare anthropic/claude-opus-4.6 openai/gpt-5.4
model-picker pick --task agent --agent opencode --limit 5
model-picker pick --agent amp --json
model-picker pick --task agent --limit 5 --contract  # stable model-picker.selection v1 envelope
```

### Export

```bash
model-picker export --format markdown --limit 10 --output ./models.md
model-picker export --format json --limit 20
model-picker export --format csv
model-picker export --format ndjson
```

### Setup and Config

```bash
model-picker onboard      # Interactive setup, saves FIRECRAWL_API_KEY to config
model-picker configure     # Re-run configuration
model-picker doctor        # Check config, env vars, and live access status
```

### Development

```bash
bun run dev:web    # Start web dashboard
bun run dev:cli   # Start CLI in watch mode
bun run dev:tui    # Start terminal UI
bun run refresh    # Refresh local model snapshot
bun run build     # Build all packages
bun run test       # Run tests
bun run typecheck  # Type check all packages
```

## Agent-Aware Picks

The `pick` command recommends models tailored to specific coding agents.

```bash
model-picker pick --task agent --agent opencode --limit 5
model-picker pick --task review --agent claude-code --limit 5
model-picker pick --agent amp --json
```

Harness adapters MUST use `pick --contract` rather than the wider, command-specific `--json` payload. The contract envelope contains `contract`, `version`, `request`, and bounded `choices` fields.

Supported agents:
- amp
- opencode
- claude-code
- codex
- cursor

When --agent is set and --task is omitted, defaults to --task agent.

## OpenRouter Filter Flags

These apply to live commands (top, get):

- --order: most-popular | top-weekly | newest
- --input-modalities: text, image, audio, video
- --output-modalities: text, image, audio, embeddings
- --categories: programming, reasoning, roleplay, image, video
- --max-price: number (per 1M input tokens)
- --zdr: top of DL, rate limited

## Environment Variables

FIRECRAWL_API_KEY    # Optional; enables Firecrawl fallback for live commands
MODEL_PICKER_CONFIG_DIR  # Override config directory

## Config File Location

- macOS/Linux: ~/.config/model-picker/config.json
- Windows: %APPDATA%\model-picker\config.json
- Override: MODEL_PICKER_CONFIG_DIR env var

## Architecture

model-picker is a Bun + Turborepo monorepo:

apps/cli/      # CAC + @clack/prompts command interface
apps/web/     # Astro + Solid web dashboard
apps/tui/     # OpenTUI + Solid terminal interface
packages/catalog/     # Snapshot loading, filtering, sorting, pick
packages/domain/     # Canonical model types and ranking
packages/presenters/  # Cross-surface row/summary formatting
packages/ui-core/     # Shared Solid view-model/state
packages/ingest/      # OpenRouter fetch, speed scraping, consolidation

See ./ARCHITECTURE.md for full system diagram.

## Quick Reference

# Discover models (live)
npx model-picker top --order most-popular --limit 10
# Compare two models (snapshot)
npx model-picker compare anthropic/claude-opus-4.6 openai/gpt-5.4

# Pick best model for an agent (snapshot)
npx model-picker pick --agent opencode --limit 5

# Export results
npx model-picker export --format markdown --limit 10 --output ./models.md

# Check setup
npx model-picker doctor

## Links

Web docs: https://model-picker.dev/docs
GitHub: https://github.com/byhow/model-picker
npm: https://www.npmjs.com/package/model-picker

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.