# ai ## Purpose Abstractions for LLM inference, provider management, tool formatting, and token counting. Decouples the core agent loop from specific API implementations. ## Key Files | File | Description | |------|-------------| | `index.ts` | Brief description of purpose | | `model-catalog-compat.ts` | Brief description of purpose | | `model-catalog.ts` | Brief description of purpose | | `model-discovery.ts` | Brief description of purpose | | `model-enrich.ts` | Brief description of purpose | | `model-manager.ts` | Brief description of purpose | | `model-picker.ts` | Brief description of purpose | | `model-registry.ts` | Brief description of purpose | | `pricing.ts` | Brief description of purpose | | `provider-registry.ts` | Brief description of purpose | | `provider-status.ts` | Brief description of purpose | | `register-providers.ts` | Brief description of purpose | | `sse.ts` | Brief description of purpose | | `types.ts` | Common interfaces for AI requests, responses, and streams | ## Subdirectories | Directory | Purpose | |-----------|---------| | `providers/` | Concrete implementations for Anthropic, OpenAI, Gemini, Ollama, etc. (see `providers/AGENTS.md`) | ## For AI Agents ### Working In This Directory - When adding a new provider, implement the standard interface defined in `types.ts` and register it in `registry.ts`. - Ensure streaming outputs are parsed reliably. - Handle rate limits (429) gracefully, exposing standard retry hints to the caller. ### Testing Requirements - Unit tests for formatting and parsing. - Integration tests (or skipped e2e tests) for actual provider connectivity. ### Common Patterns - Factory functions for instantiating provider clients. - Normalization of diverse API error structures into standard exceptions. ## Dependencies ### Internal - Consumed by `src/agent/loop.ts`. ### External - HTTP fetch (Bun.fetch).