docs: detail memory tools and local models

docs/concepts/memory.md

@@ -102,3 +102,22 @@ Local mode:
 - Set `agents.defaults.memorySearch.provider = "local"`.
 - Provide `agents.defaults.memorySearch.local.modelPath` (GGUF or `hf:` URI).
 - Optional: set `agents.defaults.memorySearch.fallback = "none"` to avoid remote fallback.
+
+### How the memory tools work
+
+- `memory_search` semantically searches Markdown chunks (~400 token target, 80-token overlap) from `MEMORY.md` + `memory/**/*.md`. It returns snippet text (capped ~700 chars), file path, line range, score, provider/model, and whether we fell back from local → remote embeddings. No full file payload is returned.
+- `memory_get` reads a specific memory Markdown file (workspace-relative), optionally from a starting line and for N lines. Paths outside `MEMORY.md` / `memory/` are rejected.
+- Both tools are enabled only when `memorySearch.enabled` resolves true for the agent.
+
+### What gets indexed (and when)
+
+- File type: Markdown only (`MEMORY.md`, `memory/**/*.md`).
+- Index storage: per-agent SQLite at `~/.clawdbot/state/memory/<agentId>.sqlite` (configurable via `agents.defaults.memorySearch.store.path`, supports `{agentId}` token).
+- Freshness: watcher on `MEMORY.md` + `memory/` marks the index dirty (debounce 1.5s). Sync runs on session start, on first search when dirty, and optionally on an interval. Reindex triggers when embedding model/provider or chunk sizes change.
+
+### Local embedding auto-download
+
+- Default local embedding model: `hf:ggml-org/embeddinggemma-300M-GGUF/embeddinggemma-300M-Q8_0.gguf` (~0.6 GB).
+- When `memorySearch.provider = "local"`, `node-llama-cpp` resolves `modelPath`; if the GGUF is missing it **auto-downloads** to the cache (or `local.modelCacheDir` if set), then loads it. Downloads resume on retry.
+- Native build requirement: run `pnpm approve-builds`, pick `node-llama-cpp`, then `pnpm rebuild node-llama-cpp`.
+- Fallback: if local setup fails and `memorySearch.fallback = "openai"`, we automatically switch to remote embeddings (`openai/text-embedding-3-small` unless overridden) and record the reason.

docs/gateway/local-models.md

@@ -55,6 +55,64 @@ Best current local stack. Load MiniMax M2.1 in LM Studio, enable the local serve
 - Adjust `contextWindow`/`maxTokens` if your LM Studio build differs.
 - For WhatsApp, stick to Responses API so only final text is sent.
 
+## Model catalog (remote + local)
+
+| Tier | Model ID | Context | Who downloads | Notes |
+| --- | --- | --- | --- | --- |
+| Hosted | `anthropic/claude-opus-4-5` | 200k | Provider | Latest Claude; keep as high-quality fallback. |
+| Hosted | `anthropic/claude-sonnet-4-5` | 200k | Provider | Cheaper Claude; good default. |
+| Hosted | `openai/gpt-4.1` | 128k | Provider | Latest GPT-4; strong tools + reasoning. |
+| Hosted | `openai/gpt-4.1-mini` | 128k | Provider | Fast/cheap GPT-4 family; good fallback. |
+| Local | `lmstudio/minimax-m2.1-gs32` | ~196k (build-dependent) | You (LM Studio UI) | Recommended local heavy model; keep loaded. |
+| Local | Custom `vllm` / `litellm` model | server-defined | You (server) | Any OpenAI-compatible endpoint; align context settings. |
+
+Keep hosted models configured even when running local; use `models.mode: "merge"` so fallbacks stay available.
+
+### Hybrid config: hosted primary, local fallback
+
+```json5
+{
+  agents: {
+    defaults: {
+      model: {
+        primary: "anthropic/claude-sonnet-4-5",
+        fallbacks: ["lmstudio/minimax-m2.1-gs32", "openai/gpt-4.1-mini"]
+      },
+      models: {
+        "anthropic/claude-sonnet-4-5": { alias: "Sonnet" },
+        "lmstudio/minimax-m2.1-gs32": { alias: "MiniMax Local" },
+        "openai/gpt-4.1-mini": { alias: "GPT-4.1 mini" }
+      }
+    }
+  },
+  models: {
+    mode: "merge",
+    providers: {
+      lmstudio: {
+        baseUrl: "http://127.0.0.1:1234/v1",
+        apiKey: "lmstudio",
+        api: "openai-responses",
+        models: [
+          {
+            id: "minimax-m2.1-gs32",
+            name: "MiniMax M2.1 GS32",
+            reasoning: false,
+            input: ["text"],
+            cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
+            contextWindow: 196608,
+            maxTokens: 8192
+          }
+        ]
+      }
+    }
+  }
+}
+```
+
+### Local-first with hosted safety net
+
+Swap the primary and fallback order; keep the same providers block and `models.mode: "merge"` so you can fall back to Sonnet or GPT-4.1 when the local box is down.
+
 ## Other OpenAI-compatible local proxies
 
 vLLM, LiteLLM, OAI-proxy, or custom gateways work if they expose an OpenAI-style `/v1` endpoint. Replace the provider block above with your endpoint and model ID: