Spec: park vision as a need of future agents, not current ones

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

docs/Architecture-Build-Spec.md

@@ -1424,9 +1424,10 @@ The rest resolve when reached, not now.
   and know nothing about which model is loaded. (Measured 2026-09-28: they only
   throttled the benchmark, and the token limiter leaked until LiteLLM restarted.)
   - **Preferred: one local model for every local role.** No swaps at all. It
-    needs one model good enough as the dev executor, for ops' Home Assistant tool
+    needs one model good enough as the dev executor and for ops' Home Assistant
-    calls, and for vision (next item). The executor benchmark and the vision and
+    tool calls. The executor benchmark and the ops evaluation decide whether one
-    ops evaluations decide whether one exists.
+    exists. More agents are planned for other jobs, which makes one shared model
+    more valuable still, and some of them may add vision (next item).
   - **Fallback: agents tag in and out.** If no single model fits, the agents
     coordinate their use of the host rather than interleave: an agent holds the
     host for a stretch of work on its model, then hands it over, so a swap happens
@@ -1434,24 +1435,14 @@ The rest resolve when reached, not now.
     groups), where VRAM allows, can reduce how often that's needed. How the
     hand-over works (a lease, a queue, or priority for ops alerts) is only worth
     designing if the preferred option fails.
-- **Open: vision (A10, A15).** Needed for `nisse-dev`'s OpenSCAD work, where the
+- **Parked: vision, for future agents.** More agents are planned for other kinds
-  model must look at a rendered part to check it against the design, and for
+  of jobs, and some will need vision: checking OpenSCAD renders against a design,
-  video-file management planned later (which agent owns it is not yet
+  or managing video files. Neither current agent needs it. The local candidates on
-  decided). The local candidates on the
+  the 4090 host already accept images, and video as sampled frames, but the agent
-  4090 host accept images, and video as sampled frames, but the agent is
+  code is text-only. Building it will need tools that return images, image content
-  text-only: canonical events carry strings, and neither renderer can send an
+  in canonical events (A10) and both renderers, storage that keeps the durable log
-  image. Needed before any vision work:
+  (A2) small, and image tokens in the context budget. How well a model sees is
-  - a way for a tool to return an image, e.g. an OpenSCAD render or a frame from
+  measured when that work starts, not before.
-    a video file;
-  - image content in the canonical event model, rendered as OpenAI image parts
-    and Anthropic image blocks, so delegation to a frontier model keeps it;
-  - storage that keeps the durable log (A2) small, e.g. a reference and hash
-    rather than the image bytes;
-  - image tokens counted in the context budget; video frames multiply them.
-  Which local model is good enough at seeing is measured separately from
-  tool calling: an evaluation with rendered OpenSCAD parts carrying seeded
-  mistakes, and sample video frames.
-
 ---
 
 ## Repository Layout