From e4d902860a9eaaeb16950c7e48151871a4be21c3 Mon Sep 17 00:00:00 2001 From: GitHub CI Date: Thu, 23 Jul 2026 03:12:34 -0700 Subject: [PATCH] =?UTF-8?q?deploy(vera):=20back=20to=20the=20local=20lane?= =?UTF-8?q?=20=E2=80=94=20smart=20is=20stable=20again?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Reverts the 2026-07-23 cloud pin. The ML node stabilised and :8041 now serves 262144 tokens (verified against the live endpoint), roughly 7x the panel's ~37k-token finder prompts — so the failure that forced the pin is gone. That was a 32768-token TOTAL ceiling (prompt + output) which every large-diff review overran, falling back to Groq's free tier and its 12000 TPM limit, exhausting the panel and leaving PRs unreviewed. Deleting the block returns to the host default (protolabs/reasoning, aliased to protolabs/smart), which is where model settings belong — they are host-scoped by ADR 0047 and the agent-layer pin only existed to outrank the box default during the outage. Applied to the live config volume and restarted: LangGraph agent initialized (model: protolabs/reasoning). Local inference also takes the panel back to zero marginal cost per review. Co-Authored-By: Claude Opus 4.8 (1M context) --- deploy/vera.langgraph-config.yaml | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/deploy/vera.langgraph-config.yaml b/deploy/vera.langgraph-config.yaml index 9312b8e..c0deb41 100644 --- a/deploy/vera.langgraph-config.yaml +++ b/deploy/vera.langgraph-config.yaml @@ -35,20 +35,6 @@ 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]