feat(llm): add Ollama provider + provider selection and CLI override; update README and logsv0.0.1

1 files changed, 34 insertions, 6 deletions

diff --git a/README.md b/README.md
index c3c77c2..1649e1b 100644
--- a/README.md
+++ b/README.md
@@ -8,14 +8,34 @@ At the moment this project is only in the proof of concept phase.
 
 ## LLM provider
 
-Hexai exposes a simple LLM provider interface and uses OpenAI by default for
-code completion when `OPENAI_API_KEY` is present in the environment.
+Hexai exposes a simple LLM provider interface. It supports OpenAI and a local
+Ollama server. Provider selection and models are configured via environment
+variables.
 
-- Required: set `OPENAI_API_KEY` to your OpenAI API key.
-- Optional: set `OPENAI_MODEL`.
-- Optional: set `OPENAI_BASE_URL` to point at a compatible endpoint.
+### Selecting a provider
 
-If no key is configured, Hexai will fall back to a basic, local completion.
+- Set `HEXAI_LLM_PROVIDER` to `openai` or `ollama` to force a provider.
+- If not set, Hexai auto‑detects:
+  - Uses OpenAI when `OPENAI_API_KEY` is present.
+  - Uses Ollama when any `OLLAMA_*` variables are present.
+  - Otherwise, Hexai falls back to a basic, local completion.
+
+### OpenAI configuration
+
+- Required: `OPENAI_API_KEY` — your OpenAI API key.
+- Optional: `OPENAI_MODEL` — model name (default: `gpt-4o-mini`).
+- Optional: `OPENAI_BASE_URL` — override the API base (e.g., a compatible endpoint).
+
+### Ollama configuration (local)
+
+- Optional: `OLLAMA_MODEL` — model name/tag (default: `qwen2.5-coder:latest`).
+- Optional: `OLLAMA_BASE_URL` or `OLLAMA_HOST` — base URL to Ollama
+  (default: `http://localhost:11434`).
+
+Notes:
+- For Ollama, ensure the model is available locally (e.g., `ollama pull qwen2.5-coder:latest`).
+- If you run Ollama in OpenAI‑compatible mode, you may alternatively use the
+  OpenAI provider with `OPENAI_BASE_URL` pointing to your local endpoint.
 
 ## CLI usage and configuration
 
@@ -31,6 +51,7 @@ If no key is configured, Hexai will fall back to a basic, local completion.
     - `always-full`: always include the full file (may be slower/costly)
   - `-context-window-lines`: line count for the sliding window when `context-mode=window`.
   - `-max-context-tokens`: budget for additional context tokens. If the flag isn’t provided, `HEXAI_MAX_CONTEXT_TOKENS` is used when set.
+  - `-provider`: LLM provider override: `openai` or `ollama` (overrides `HEXAI_LLM_PROVIDER`).
 
 Notes:
 - Token estimation for truncation uses a simple 4 chars/token heuristic.
@@ -48,3 +69,10 @@ Notes:
 | `-max-context-tokens`   | `HEXAI_MAX_CONTEXT_TOKENS` | Token budget for additional context.               |
 | `-log-preview-limit`    | `HEXAI_LOG_PREVIEW_LIMIT`  | Limit characters shown in LLM preview logs.        |
 | `-no-disk-io`           | `HEXAI_NO_DISK_IO`         | Disallow any disk reads for context.               |
+| `-provider`             | `HEXAI_LLM_PROVIDER`       | Force LLM provider: `openai` or `ollama`.          |
+
+### Environment quick reference (providers)
+
+- `HEXAI_LLM_PROVIDER`: `openai` | `ollama` (optional; otherwise auto‑detect).
+- OpenAI: `OPENAI_API_KEY` (required), `OPENAI_MODEL`, `OPENAI_BASE_URL`.
+- Ollama: `OLLAMA_MODEL`, `OLLAMA_BASE_URL` or `OLLAMA_HOST`.