chore(model): remove inert --name flag from obol model setup custom

.agents/skills/obol-stack-dev/references/llm-routing.md

@@ -53,7 +53,6 @@ obol model remove qwen3.5:9b
 obol model remove qwen3.5:4b
 
 obol model setup custom \
-    --name spark1-vllm \
     --endpoint http://192.168.18.23:8000/v1 \
     --model qwen36-deep
 # `setup custom` validates the endpoint, patches LiteLLM, and internally calls
@@ -64,7 +63,7 @@ obol model list      # confirm the custom entry is the only local model
 obol model status    # provider state
 ```
 
-The flow scripts (`flows/lib.sh::route_llm_via_obol_cli`) wrap this exact sequence behind `OBOL_LLM_ENDPOINT` / `OBOL_LLM_MODEL` / `OBOL_LLM_NAME` / `OBOL_LLM_API_KEY` env vars so smoke tests target a GPU host without burning host CPU on local Ollama.
+The flow scripts (`flows/lib.sh::route_llm_via_obol_cli`) wrap this exact sequence behind `OBOL_LLM_ENDPOINT` / `OBOL_LLM_MODEL` / `OBOL_LLM_API_KEY` env vars so smoke tests target a GPU host without burning host CPU on local Ollama.
 
 ## Paid Routing (`paid/<remote-model>`)
 

CLAUDE.md

@@ -244,7 +244,6 @@ obol model remove qwen3.5:9b
 obol model remove qwen3.5:4b
 
 obol model setup custom \
-    --name spark1-vllm \
     --endpoint http://192.168.18.23:8000/v1 \
     --model qwen36-deep
 # `setup custom` validates the endpoint, patches LiteLLM, and internally calls
@@ -259,7 +258,7 @@ obol model list                                                # confirm head of
 obol model status                                              # show provider state
 ```
 
-The flow scripts (`flows/lib.sh:route_llm_via_obol_cli`) wrap this exact sequence behind `OBOL_LLM_ENDPOINT` / `OBOL_LLM_MODEL` / `OBOL_LLM_NAME` / `OBOL_LLM_API_KEY` env vars, so smoke tests can target a GPU host without burning host CPU on local Ollama.
+The flow scripts (`flows/lib.sh:route_llm_via_obol_cli`) wrap this exact sequence behind `OBOL_LLM_ENDPOINT` / `OBOL_LLM_MODEL` / `OBOL_LLM_API_KEY` env vars, so smoke tests can target a GPU host without burning host CPU on local Ollama.
 
 **Per-instance overlay**: `buildLiteLLMRoutedOverlay()` reuses "ollama" provider slot pointing at `litellm.llm.svc:4000/v1` with `api: openai-completions`. App → litellm:4000 → routes by model name → actual API.
 

cmd/obol/model.go

@@ -264,20 +264,18 @@ func modelSetupCustomCommand(cfg *config.Config) *cli.Command {
 		Name:  "custom",
 		Usage: "Add a custom OpenAI-compatible endpoint (validates before adding)",
 		Flags: []cli.Flag{
-			&cli.StringFlag{Name: "name", Usage: "Short label for the endpoint (informational only — LiteLLM keys the route by --model, not --name)", Required: true},
 			&cli.StringFlag{Name: "endpoint", Usage: "Full base URL (e.g. http://host:8000/v1)", Required: true},
 			&cli.StringFlag{Name: "model", Usage: "Model identifier at the endpoint — this is also the LiteLLM model_name the agent will call", Required: true},
 			&cli.StringFlag{Name: "api-key", Usage: "API key (optional, some endpoints don't require it)"},
 			&cli.BoolFlag{Name: "no-sync", Usage: "Skip the agent model sync (batch with other model commands, then run `obol model sync` once)"},
 		},
 		Action: func(ctx context.Context, cmd *cli.Command) error {
 			u := getUI(cmd)
-			name := cmd.String("name")
 			endpoint := cmd.String("endpoint")
 			modelName := cmd.String("model")
 			apiKey := cmd.String("api-key")
 
-			if err := model.AddCustomEndpoint(cfg, u, name, endpoint, modelName, apiKey); err != nil {
+			if err := model.AddCustomEndpoint(cfg, u, endpoint, modelName, apiKey); err != nil {
 				return err
 			}
 

flows/buy-external.sh

@@ -61,7 +61,6 @@
 #   EXTERNAL_LOG_BLOCKS_BACK    default: 30 (~6 min on Base Sepolia at 2s/blk)
 #   OBOL_LLM_ENDPOINT           default: http://127.0.0.1:8000/v1
 #   OBOL_LLM_MODEL              default: qwen36-deep (27B-class)
-#   OBOL_LLM_NAME               default: external-llm
 #
 # Exit code: 0 on PASS (every step pass), 1 on any FAIL.
 
@@ -107,7 +106,6 @@ EXTERNAL_LOG_BLOCKS_BACK="${EXTERNAL_LOG_BLOCKS_BACK:-30}"
 
 OBOL_LLM_ENDPOINT="${OBOL_LLM_ENDPOINT:-http://127.0.0.1:8000/v1}"
 OBOL_LLM_MODEL="${OBOL_LLM_MODEL:-qwen36-deep}"
-OBOL_LLM_NAME="${OBOL_LLM_NAME:-external-llm}"
 
 # Resolve OBOL_ROOT before sourcing helpers — lib.sh re-derives it but
 # operating on the canonical path simplifies later relative paths.
@@ -449,7 +447,7 @@ detect_buyer_runtime bob
 # ─────────────────────────────────────────────────────────────────
 # STEP 5: Repoint LiteLLM at OBOL_LLM_ENDPOINT and add the live RPC route
 # ─────────────────────────────────────────────────────────────────
-step "Bob: route LiteLLM via $OBOL_LLM_NAME ($OBOL_LLM_MODEL)"
+step "Bob: route LiteLLM via $OBOL_LLM_MODEL ($OBOL_LLM_ENDPOINT)"
 if route_llm_via_obol_cli bob; then
     pass "LiteLLM routed via $OBOL_LLM_ENDPOINT"
 else

flows/lib.sh

@@ -572,12 +572,10 @@ bootstrap_flow_workspace() {
 # OBOL_LLM_MODEL is the upstream model id (default qwen36-deep, 27B-class).
 # qwen36-fast (4B) is faster but flakes on long single-shot agent prompts; see
 # the flow-13/14 step 46 retry-wrapper rationale in lib-dual-stack.sh.
-# OBOL_LLM_NAME is the LiteLLM short name registered for the endpoint (default
-# external-llm).
 #
 # Sequence (all model edits use --no-sync so we trigger only one Hermes
 # helmfile rollout at the end):
-#   1. obol model setup custom --name … --endpoint … --model … --no-sync
+#   1. obol model setup custom --endpoint … --model … --no-sync
 #      (validates the endpoint, patches LiteLLM, hot-adds the model.)
 #   2. obol model prefer <model> --no-sync
 #      (configured LiteLLM order is the primary-model contract.)
@@ -587,13 +585,12 @@ bootstrap_flow_workspace() {
 # Each peer (alice/bob) routes independently — caller passes the runner.
 route_llm_via_obol_cli() {
     local runner=$1
-    local model name
+    local model
 
     if [ -n "${OBOL_LLM_ENDPOINT:-}" ]; then
         model="${OBOL_LLM_MODEL:-qwen36-deep}"
-        name="${OBOL_LLM_NAME:-external-llm}"
 
-        local args=(model setup custom --no-sync --name "$name" --endpoint "$OBOL_LLM_ENDPOINT" --model "$model")
+        local args=(model setup custom --no-sync --endpoint "$OBOL_LLM_ENDPOINT" --model "$model")
         if [ -n "${OBOL_LLM_API_KEY:-}" ]; then
             args+=(--api-key "$OBOL_LLM_API_KEY")
         fi

internal/embed/skills/monetize-guide/SKILL.md

@@ -143,7 +143,7 @@ Two steps: first bridge the endpoint into LiteLLM, then sell LiteLLM.
 
 ```bash
 # Step A: Add the external endpoint to LiteLLM
-obol model setup custom --name <name> \
+obol model setup custom \
   --endpoint <full_url_with_v1> \
   --model "<model_name_at_endpoint>"
 

internal/model/model.go

@@ -804,13 +804,11 @@ func RemoveModel(cfg *config.Config, u *ui.UI, modelName string) error {
 // because the agent then strips it and calls LiteLLM with a key that doesn't
 // match.
 //
-// The `name` arg is informational only. It is surfaced via
-// `obol model status` / `list` for human reference but does NOT participate
-// in the LiteLLM route key. Two custom endpoints that publish the same
-// `modelName` will overwrite each other in the LiteLLM ConfigMap; that is
-// the natural "repoint my model" behavior an operator running
-// `obol model setup custom` wants when they re-run the command.
-func AddCustomEndpoint(cfg *config.Config, u *ui.UI, name, endpoint, modelName, apiKey string) error {
+// Two custom endpoints that publish the same `modelName` will overwrite
+// each other in the LiteLLM ConfigMap; that is the natural "repoint my
+// model" behavior an operator running `obol model setup custom` wants when
+// they re-run the command.
+func AddCustomEndpoint(cfg *config.Config, u *ui.UI, endpoint, modelName, apiKey string) error {
 	kubectlBinary := filepath.Join(cfg.BinDir, "kubectl")
 	kubeconfigPath := filepath.Join(cfg.ConfigDir, "kubeconfig.yaml")
 
@@ -840,14 +838,7 @@ func AddCustomEndpoint(cfg *config.Config, u *ui.UI, name, endpoint, modelName,
 
 	entry := buildCustomEndpointEntry(modelName, clusterEndpoint, apiKey)
 
-	// Patch ConfigMap for persistence. The display label is logged so an
-	// operator can correlate the call with their `--name` arg, but it isn't
-	// part of the route key.
-	if name != "" {
-		u.Infof("Adding custom endpoint %q (model: %s) to LiteLLM config", name, modelName)
-	} else {
-		u.Infof("Adding custom endpoint (model: %s) to LiteLLM config", modelName)
-	}
+	u.Infof("Adding custom endpoint (model: %s) to LiteLLM config", modelName)
 
 	if err := patchLiteLLMConfig(kubectlBinary, kubeconfigPath, []ModelEntry{entry}); err != nil {
 		return fmt.Errorf("failed to update LiteLLM config: %w", err)
@@ -856,10 +847,10 @@ func AddCustomEndpoint(cfg *config.Config, u *ui.UI, name, endpoint, modelName,
 	// Hot-add via API (no restart needed).
 	if err := hotAddModels(cfg, u, []ModelEntry{entry}); err != nil {
 		u.Warnf("Hot-add failed, falling back to restart: %v", err)
-		return RestartLiteLLM(cfg, u, name)
+		return RestartLiteLLM(cfg, u, modelName)
 	}
 
-	u.Successf("Custom endpoint %q added (model: %s)", name, modelName)
+	u.Successf("Custom endpoint added (model: %s)", modelName)
 
 	return nil
 }