From 673b7bcffc67a829264e4326bbb88fcba6000c2c Mon Sep 17 00:00:00 2001
From: Robert Helewka <r@helu.ca>
Date: Tue, 12 May 2026 15:19:36 -0400
Subject: [PATCH] Validator FastAgent Config updates

---
 validator/fastagent.config.yaml          | 11 +++++++----
 validator/fastagent.secrets.yaml.example |  9 +++++----
 2 files changed, 12 insertions(+), 8 deletions(-)

diff --git a/validator/fastagent.config.yaml b/validator/fastagent.config.yaml
index b0b07f5..291c1fc 100644
--- a/validator/fastagent.config.yaml
+++ b/validator/fastagent.config.yaml
@@ -3,9 +3,12 @@
 # Secrets (api_key, MCP bearer tokens) live in fastagent.secrets.yaml
 # (gitignored) and merge with this file at runtime.
 
-# Local llama.cpp on Nyx (OpenAI-compatible). Override via
-# fastagent.secrets.yaml if you want to point at a different model server.
-default_model: openai.Qwen3.5-35B-A3B-UD-Q4_K_XL.gguf
+# Local llama.cpp on Nyx — served via fast-agent's Generic (OpenAI-compatible)
+# provider. The `openai` provider slot is reserved for cloud OpenAI endpoints
+# (including AWS Bedrock Mantle's Chat Completions path — see
+# pallas/docs/bedrock.md). Override via fastagent.secrets.yaml if you want to
+# point at a different model server.
+default_model: generic.Qwen3.5-35B-A3B-UD-Q4_K_XL.gguf
 
 # Capabilities for the model — Pallas registers it with fast-agent's
 # ModelDatabase using these values. vision: true so we can validate image
@@ -16,7 +19,7 @@ model_capabilities:
   max_output_tokens: 16384
 
 # ── LLM Providers ───────────────────────────────────────────────────────────
-openai:
+generic:
   base_url: "http://nyx.helu.ca:22079/v1"
 
 # ── MCP Servers ─────────────────────────────────────────────────────────────
diff --git a/validator/fastagent.secrets.yaml.example b/validator/fastagent.secrets.yaml.example
index 8f68481..71ad7c7 100644
--- a/validator/fastagent.secrets.yaml.example
+++ b/validator/fastagent.secrets.yaml.example
@@ -4,10 +4,11 @@
 # gitignored; the .yaml.example is committed.
 
 # ── LLM provider keys ───────────────────────────────────────────────────────
-# Local llama.cpp doesn't authenticate, but fast-agent requires the key field
-# to be present. "0000" or any non-empty string is fine.
-openai:
-  api_key: "0000"
+# Local llama.cpp on Nyx is served via fast-agent's Generic provider
+# (fastagent.config.yaml: `default_model: generic.…`). The Generic provider
+# injects "ollama" as a placeholder key automatically, so no explicit entry
+# is needed here. If you later point `default_model` at a cloud OpenAI model,
+# add an `openai.api_key` stanza.
 
 # ── MCP server bearer tokens ────────────────────────────────────────────────
 mcp: