From a7f72cf1fcc0768fdfc6b5fe2fe762b76d89434c Mon Sep 17 00:00:00 2001 From: GitHub CI Date: Thu, 23 Jul 2026 01:37:18 -0700 Subject: [PATCH] deploy(vera): pin the cloud model while smart is pulled MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit protoAgent's default model_name is `protolabs/reasoning`, which the gateway aliases to `protolabs/smart` — the local :8041 lane. With smart being pulled, an unpinned Vera goes down with it. `protolabs/cloud` -> deepseek-v4-pro (1M context). That also clears the failure that exhausted the panel this morning: the local lane served 32768 tokens TOTAL (prompt + output, verified empirically) against ~36.8k-token finder prompts, so large-diff reviews failed on context length after the Groq fallback failed on TPM. Model settings are host-scoped by default (ADR 0047); this agent-layer pin deliberately outranks the box default and should be deleted once smart returns. Applied to the LIVE config volume as well — the seed is copy-once and would otherwise only reach a fresh instance. Note: the gateway has no context profile for this alias, so compaction degrades from 'fraction:0.8' to 'messages:60'. Harmless at 1M context, worth knowing. Co-Authored-By: Claude Opus 4.8 (1M context) --- deploy/vera.langgraph-config.yaml | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/deploy/vera.langgraph-config.yaml b/deploy/vera.langgraph-config.yaml index c0deb41..9312b8e 100644 --- a/deploy/vera.langgraph-config.yaml +++ b/deploy/vera.langgraph-config.yaml @@ -35,6 +35,20 @@ a2a: - "review https://github.com/protoLabsAI/protoAgent/pull/64" - "what's your verdict on PR #42 in protoLabsAI/qaEngineer?" +# Model — pinned to the CLOUD lane 2026-07-23 while the local `smart` backend +# (:8041) is pulled. Model settings are normally HOST-scoped (ADR 0047) and this +# agent layer deliberately outranks the box default, because the default +# `protolabs/reasoning` is a gateway ALIAS to `protolabs/smart` — leaving it unset +# would take the review panel down with the backend it points at. +# +# protolabs/cloud -> deepseek-v4-pro: 1M context, so the panel's ~37k-token finder +# prompts fit with room to spare (the previous local lane capped at 32k total, +# which is what exhausted the panel on large diffs). +# +# Revert to the host default (delete this block) once smart is back. +model: + name: protolabs/cloud + plugins: enabled: [workflows, github, pr-reviewer]