summary system backend implementation

docs/services/orchestration-service.md

@@ -48,7 +48,7 @@ src/
 ├── routes/
 │   ├── chat.js        # POST /chat and POST /chat/stream
 │   ├── sessions.js    # Session CRUD proxy
-│   ├── projects.js    # Project CRUD proxy
+│   ├── projects.js    # Project CRUD proxy — passes req.body straight through
 │   ├── episodes.js    # Episode list and delete proxy
 │   ├── settings.js    # GET /settings and PATCH /settings
 │   ├── health.js      # GET /health — pings all four services
@@ -75,6 +75,7 @@ via `appSettings.load()` — changes apply immediately without a service restart
 | `repeatPenalty` | 1.1 | Repeat token penalty |
 | `topP` | 0.9 | Nucleus sampling probability mass |
 | `topK` | 40 | Top-K token candidates per step |
+| `systemPrompt` | *(ORCHESTRATION.SYSTEM_PROMPT)* | Global system prompt. `null` reverts to hardcoded constant. |
 
 Defaults are defined in `config/settings.js` and fall back to constants in
 `@nexusai/shared`. Values saved in `settings.json` take precedence.
@@ -91,41 +92,43 @@ difference is how the inference response is delivered to the client.
    step needed.
 
 2. **Project context resolution** — if the session has a `project_id`, fetch
-   the project and all its session IDs. Used to scope semantic search. See
-   `memory-isolation.md` for full behaviour.
+   the project and all its session IDs. Used to scope semantic search. The
+   project's `system_prompt` is also read at this step if set.
 
-3. **Recent episode retrieval** — fetch the most recent episodes for the
+3. **System prompt resolution** — three-tier hierarchy:
+   - `project.system_prompt` — if the session is in a project and it's set (highest priority)
+   - `settings.systemPrompt` — global setting from `settings.json`
+   - `ORCHESTRATION.SYSTEM_PROMPT` — hardcoded constant in `@nexusai/shared` (last resort)
+
+4. **Recent episode retrieval** — fetch the most recent episodes for the
    session (`recentEpisodeLimit`, default 5).
 
-4. **Semantic search** — embed the user message, query Qdrant for the top
+5. **Semantic search** — embed the user message, query Qdrant for the top
    most similar past episodes (`semanticLimit`, `scoreThreshold`). Deduplicated
    against recent episodes. Non-critical — if it fails, pipeline continues with
    recency-only context.
 
-5. **Entity search** — reuse the embedded user message vector to query the
-   `entities` Qdrant collection (score threshold 0.6, limit 5). Returns
-   entity payloads (`name`, `type`, `notes`) directly — no SQLite roundtrip
-   needed. Non-critical — if it fails, pipeline continues without entity context.
+6. **Entity search** — query the `entities` Qdrant collection filtered by
+   `projectId`. Non-project sessions receive no entity context. Non-critical.
 
-6. **Prompt assembly** — combine system prompt, entity context, semantic
-   episodes, recent episodes, and user message.
+7. **Prompt assembly** — combine resolved system prompt, entity context,
+   semantic episodes, recent episodes, and user message.
 
-7. **Inference** — send to inference service with settings-derived parameters
+8. **Inference** — send to inference service with settings-derived parameters
    (temperature, topP, topK, repeatPenalty). `/chat` awaits full response;
    `/chat/stream` pipes SSE chunks to the client.
 
-8. **Episode write** — write the exchange back to memory. Fire-and-forget
-   for `/chat`; awaited for `/chat/stream` to ensure the full text is
-   accumulated before saving.
+9. **Episode write** — write the exchange back to memory with `projectId`.
+   Fire-and-forget for `/chat`; awaited for `/chat/stream`.
 
-9. **Auto-naming** — on `isFirstMessage && !session.name`, fire a secondary
-   inference call with a naming prompt (max 20 tokens, temperature 0.3) and
-   write the result back as `session.name`. Fully fire-and-forget.
+10. **Auto-naming** — on `isFirstMessage && !session.name`, fire a secondary
+    inference call with a naming prompt (max 20 tokens, temperature 0.3) and
+    write the result back as `session.name`. Fully fire-and-forget.
 
 ### Prompt Structure
 
 ```
-[System prompt]
+[Resolved system prompt]
 
 Here is what you know about entities relevant to this conversation:
 - {name} ({type}): {notes}
@@ -175,9 +178,9 @@ is terminated by `res.end()` after the done event.
 folder for richer metadata (label, description). Returns file size in GB.
 
 `GET /models/props` fetches directly from llama-server via `LLAMA_SERVER_URL`.
-Returns `{ contextWindow, modelAlias }`. Used by the client to display
-read-only context window size and the currently loaded model in the settings
-panel. Returns `503` if llama-server is unreachable.
+Returns `{ contextWindow, modelAlias }`. `n_ctx` is at
+`data.default_generation_settings.n_ctx` in the llama-server response.
+Returns `503` if llama-server is unreachable.
 
 ## Sessions Route Behaviour