polish: address self-review feedback

  • Remove unused isUrl() helper in /local parser (replaced by
    looksLikeUrl during earlier fix; never deleted).

  • Rename env var CODEBUFF_LOCAL_MODEL → CODEBUFF_PROVIDER_MODEL
    so all three custom-provider env vars share the CODEBUFF_(PROVIDER_)?
    prefix consistently. Clarify in JSDoc that the override is skipped
    when an agent declares its own providerOptions.baseUrl.

  • Default apiKey placeholder "codebuff" → "unused" in
    createCustomProviderModel. The literal string "codebuff" invited the
    wrong mental model (could read as "send my Codebuff key"); "unused"
    plus a comment makes the intent obvious. Local runtimes ignore the
    Authorization header entirely; we never send the user's real key on
    the direct path.

  • Extract maxRetries: 1 into CUSTOM_PROVIDER_MAX_RETRIES with a
    JSDoc explaining the choice (one retry for cold-start; more wouldn't
    help with deterministic local failures).

  • Simplify the precedence ladder in promptAiSdkStream — replace the
    nested ternary that paired apiKey-with-winning-baseUrl with a small
    sources array + .find(). Same behavior, easier to read at a glance.

Tests updated for the env var rename. All 3245 tests across CLI, SDK,
and agent-runtime still pass.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

Author: vraj00222

cli/src/commands/__tests__/local-provider.test.ts

@@ -153,10 +153,10 @@ describe('applyLocalAction (side effects on process.env)', () => {
   beforeEach(() => {
     originalBaseUrl = process.env.CODEBUFF_BASE_URL
     originalApiKey = process.env.CODEBUFF_PROVIDER_API_KEY
-    originalModel = process.env.CODEBUFF_LOCAL_MODEL
+    originalModel = process.env.CODEBUFF_PROVIDER_MODEL
     delete process.env.CODEBUFF_BASE_URL
     delete process.env.CODEBUFF_PROVIDER_API_KEY
-    delete process.env.CODEBUFF_LOCAL_MODEL
+    delete process.env.CODEBUFF_PROVIDER_MODEL
   })
 
   afterEach(() => {
@@ -165,18 +165,18 @@ describe('applyLocalAction (side effects on process.env)', () => {
     if (originalApiKey === undefined)
       delete process.env.CODEBUFF_PROVIDER_API_KEY
     else process.env.CODEBUFF_PROVIDER_API_KEY = originalApiKey
-    if (originalModel === undefined) delete process.env.CODEBUFF_LOCAL_MODEL
-    else process.env.CODEBUFF_LOCAL_MODEL = originalModel
+    if (originalModel === undefined) delete process.env.CODEBUFF_PROVIDER_MODEL
+    else process.env.CODEBUFF_PROVIDER_MODEL = originalModel
   })
 
   test('enable without model sets baseUrl, clears any previous model override', async () => {
-    process.env.CODEBUFF_LOCAL_MODEL = 'stale-model'
+    process.env.CODEBUFF_PROVIDER_MODEL = 'stale-model'
     const msg = await applyLocalAction({
       kind: 'enable',
       baseUrl: 'http://localhost:11434/v1',
     })
     expect(process.env.CODEBUFF_BASE_URL).toBe('http://localhost:11434/v1')
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBeUndefined()
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBeUndefined()
     expect(msg).toContain('ON')
     expect(msg).toContain('No model override')
     expect(msg).toContain('llama3.1:8b')
@@ -189,7 +189,7 @@ describe('applyLocalAction (side effects on process.env)', () => {
       model: 'llama3.1:8b',
     })
     expect(process.env.CODEBUFF_BASE_URL).toBe('http://localhost:11434/v1')
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBe('llama3.1:8b')
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBe('llama3.1:8b')
     expect(msg).toContain('Model override: llama3.1:8b')
   })
 
@@ -198,7 +198,7 @@ describe('applyLocalAction (side effects on process.env)', () => {
       kind: 'set-model',
       model: 'llama3.1:8b',
     })
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBeUndefined()
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBeUndefined()
     expect(msg).toContain('OFF')
   })
 
@@ -208,16 +208,16 @@ describe('applyLocalAction (side effects on process.env)', () => {
       kind: 'set-model',
       model: 'llama3.1:8b',
     })
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBe('llama3.1:8b')
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBe('llama3.1:8b')
     expect(msg).toContain('Model override: llama3.1:8b')
   })
 
   test('clear-model removes only the model, keeps baseUrl', async () => {
     process.env.CODEBUFF_BASE_URL = 'http://localhost:11434/v1'
-    process.env.CODEBUFF_LOCAL_MODEL = 'llama3.1:8b'
+    process.env.CODEBUFF_PROVIDER_MODEL = 'llama3.1:8b'
     const msg = await applyLocalAction({ kind: 'clear-model' })
     expect(process.env.CODEBUFF_BASE_URL).toBe('http://localhost:11434/v1')
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBeUndefined()
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBeUndefined()
     expect(msg).toContain('cleared')
   })
 
@@ -229,11 +229,11 @@ describe('applyLocalAction (side effects on process.env)', () => {
   test('disable clears baseUrl, apiKey, and model', async () => {
     process.env.CODEBUFF_BASE_URL = 'http://localhost:11434/v1'
     process.env.CODEBUFF_PROVIDER_API_KEY = 'ollama'
-    process.env.CODEBUFF_LOCAL_MODEL = 'llama3.1:8b'
+    process.env.CODEBUFF_PROVIDER_MODEL = 'llama3.1:8b'
     const msg = await applyLocalAction({ kind: 'disable' })
     expect(process.env.CODEBUFF_BASE_URL).toBeUndefined()
     expect(process.env.CODEBUFF_PROVIDER_API_KEY).toBeUndefined()
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBeUndefined()
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBeUndefined()
     expect(msg).toContain('OFF')
     expect(msg).toContain('llama3.1:8b')
   })
@@ -251,7 +251,7 @@ describe('applyLocalAction (side effects on process.env)', () => {
 
   test('status when on with model shows both URL and model', async () => {
     process.env.CODEBUFF_BASE_URL = 'http://localhost:1234/v1'
-    process.env.CODEBUFF_LOCAL_MODEL = 'llama3.1:8b'
+    process.env.CODEBUFF_PROVIDER_MODEL = 'llama3.1:8b'
     const msg = await applyLocalAction({ kind: 'status' })
     expect(msg).toContain('ON')
     expect(msg).toContain('http://localhost:1234/v1')
@@ -285,48 +285,48 @@ describe('parseLocalArgs + applyLocalAction end-to-end', () => {
 
   beforeEach(() => {
     originalBaseUrl = process.env.CODEBUFF_BASE_URL
-    originalModel = process.env.CODEBUFF_LOCAL_MODEL
+    originalModel = process.env.CODEBUFF_PROVIDER_MODEL
     delete process.env.CODEBUFF_BASE_URL
-    delete process.env.CODEBUFF_LOCAL_MODEL
+    delete process.env.CODEBUFF_PROVIDER_MODEL
   })
 
   afterEach(() => {
     if (originalBaseUrl === undefined) delete process.env.CODEBUFF_BASE_URL
     else process.env.CODEBUFF_BASE_URL = originalBaseUrl
-    if (originalModel === undefined) delete process.env.CODEBUFF_LOCAL_MODEL
-    else process.env.CODEBUFF_LOCAL_MODEL = originalModel
+    if (originalModel === undefined) delete process.env.CODEBUFF_PROVIDER_MODEL
+    else process.env.CODEBUFF_PROVIDER_MODEL = originalModel
   })
 
   test('user types `/local on llama3.1:8b` → URL default + model set', async () => {
     await applyLocalAction(parseLocalArgs('on llama3.1:8b'))
     expect(process.env.CODEBUFF_BASE_URL).toBe(DEFAULT_LOCAL_BASE_URL)
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBe('llama3.1:8b')
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBe('llama3.1:8b')
   })
 
   test('user types `/local llama3.1:8b` (no `on`) → same effect', async () => {
     await applyLocalAction(parseLocalArgs('llama3.1:8b'))
     expect(process.env.CODEBUFF_BASE_URL).toBe(DEFAULT_LOCAL_BASE_URL)
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBe('llama3.1:8b')
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBe('llama3.1:8b')
   })
 
   test('user types `/local on http://x/v1 llama3.1:8b` → both set', async () => {
     await applyLocalAction(parseLocalArgs('on http://x.example.com:9999/v1 llama3.1:8b'))
     expect(process.env.CODEBUFF_BASE_URL).toBe('http://x.example.com:9999/v1')
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBe('llama3.1:8b')
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBe('llama3.1:8b')
   })
 
   test('user types `/local model llama3.1:8b` after `/local on` → model added', async () => {
     await applyLocalAction(parseLocalArgs('on'))
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBeUndefined()
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBeUndefined()
     await applyLocalAction(parseLocalArgs('model llama3.1:8b'))
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBe('llama3.1:8b')
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBe('llama3.1:8b')
   })
 
   test('user types `/local off` → both cleared', async () => {
     await applyLocalAction(parseLocalArgs('on llama3.1:8b'))
     await applyLocalAction(parseLocalArgs('off'))
     expect(process.env.CODEBUFF_BASE_URL).toBeUndefined()
-    expect(process.env.CODEBUFF_LOCAL_MODEL).toBeUndefined()
+    expect(process.env.CODEBUFF_PROVIDER_MODEL).toBeUndefined()
   })
 
   test('mutations are visible via getter functions', async () => {

cli/src/commands/local-provider.ts

@@ -42,10 +42,6 @@ export type LocalCommandAction =
   | { kind: 'disable' }
   | { kind: 'invalid'; reason: string }
 
-function isUrl(token: string): boolean {
-  return token.startsWith('http://') || token.startsWith('https://')
-}
-
 function looksLikeUrl(token: string): boolean {
   // Anything with a scheme separator — caller validates the actual scheme.
   return token.includes('://')

common/src/constants/custom-provider.ts

@@ -9,5 +9,9 @@ export const PROVIDER_API_KEY_ENV_VAR = 'CODEBUFF_PROVIDER_API_KEY'
 /** Env var overriding the agent's declared model when a custom provider is active.
  *  Used by `/local on <model>` to substitute the cloud model (e.g.
  *  `anthropic/claude-opus-4-7`) with a model the local provider actually has
- *  (e.g. `llama3.1:8b`). Only takes effect when PROVIDER_BASE_URL_ENV_VAR is set. */
-export const PROVIDER_MODEL_ENV_VAR = 'CODEBUFF_LOCAL_MODEL'
+ *  (e.g. `llama3.1:8b`).
+ *
+ *  Only takes effect when PROVIDER_BASE_URL_ENV_VAR is set AND the agent
+ *  itself doesn't declare its own `providerOptions.baseUrl` — agents with an
+ *  explicit baseUrl are assumed to declare a matching model. */
+export const PROVIDER_MODEL_ENV_VAR = 'CODEBUFF_PROVIDER_MODEL'

sdk/src/impl/llm.ts

@@ -135,6 +135,14 @@ type OpenRouterUsageAccounting = {
   }
 }
 
+/**
+ * Retry count for direct calls to a custom OpenAI-compatible provider.
+ * One retry absorbs brief model-load stalls on first call. We deliberately
+ * don't retry more — local failures are usually deterministic (provider down,
+ * wrong URL, model not pulled) and extra retries only make errors slower.
+ */
+const CUSTOM_PROVIDER_MAX_RETRIES = 1
+
 /**
  * Wrap raw errors from a custom OpenAI-compatible endpoint in a friendly,
  * actionable message. Distinguishes connection failures (provider down,
@@ -366,29 +374,25 @@ export async function* promptAiSdkStream(
   }
 
   // Resolve custom-provider precedence: agent > client option > env.
-  // apiKey is paired with whichever URL "wins" to avoid mixing sources.
+  // First non-empty baseUrl wins; its apiKey comes along to avoid mixing
+  // credentials with the wrong endpoint.
+  const customSources = [
+    params.agentProviderOptions,
+    params.clientCustomProvider,
+    {
+      baseUrl: getCustomProviderBaseUrlFromEnv(),
+      apiKey: getCustomProviderApiKeyFromEnv(),
+    },
+  ]
+  const winningSource = customSources.find((s) => s?.baseUrl)
+  const resolvedBaseUrl = winningSource?.baseUrl
+  const resolvedApiKey = winningSource?.apiKey
+
+  // Model override: substitute the agent's declared model with the env-configured
+  // local model when the custom provider is active. Skipped when an agent
+  // explicitly sets its own providerOptions.baseUrl — that agent is assumed to
+  // have declared a matching model. See PROVIDER_MODEL_ENV_VAR JSDoc.
   const agentBaseUrl = params.agentProviderOptions?.baseUrl
-  const agentApiKey = params.agentProviderOptions?.apiKey
-  const clientBaseUrl = params.clientCustomProvider?.baseUrl
-  const clientApiKey = params.clientCustomProvider?.apiKey
-  const envBaseUrl = getCustomProviderBaseUrlFromEnv()
-  const envApiKey = getCustomProviderApiKeyFromEnv()
-
-  const resolvedBaseUrl = agentBaseUrl ?? clientBaseUrl ?? envBaseUrl
-  const resolvedApiKey = agentBaseUrl
-    ? agentApiKey
-    : clientBaseUrl
-      ? clientApiKey
-      : envBaseUrl
-        ? envApiKey
-        : undefined
-
-  // Model override: when a custom provider is active and CODEBUFF_LOCAL_MODEL
-  // is set, substitute the agent's declared model (which is typically a cloud
-  // model id like 'anthropic/claude-opus-4-7' that a local provider won't
-  // recognize) with the configured local model (e.g. 'llama3.1:8b').
-  // Only applies to envBaseUrl/clientBaseUrl paths — an agent that explicitly
-  // sets providerOptions.baseUrl is assumed to also have set a matching model.
   const envModelOverride =
     resolvedBaseUrl && !agentBaseUrl
       ? getCustomProviderModelFromEnv()
@@ -438,10 +442,9 @@ export async function* promptAiSdkStream(
     model: aiSDKModel,
     messages: convertCbToModelMessages(params),
     // ChatGPT OAuth: no retries (we fall back to Codebuff on first failure).
-    // Custom provider: one retry to handle brief model-load stalls without
-    // dragging out errors when the provider is actually down.
+    // Custom provider: see CUSTOM_PROVIDER_MAX_RETRIES.
     ...(isChatGptOAuth ? { maxRetries: 0 } : {}),
-    ...(isCustomProvider ? { maxRetries: 1 } : {}),
+    ...(isCustomProvider ? { maxRetries: CUSTOM_PROVIDER_MAX_RETRIES } : {}),
     // Direct routes (ChatGPT OAuth, custom provider): skip codebuff_metadata
     // and OpenRouter routing keys — neither belongs in those request bodies.
     ...(isChatGptOAuth || isCustomProvider

sdk/src/impl/model-provider.ts

@@ -296,7 +296,10 @@ function createCustomProviderModel(params: {
     provider: 'custom',
     url: ({ path: endpoint }) => `${trimmedBase}${endpoint}`,
     headers: () => ({
-      Authorization: `Bearer ${apiKey ?? 'codebuff'}`,
+      // Most local runtimes (Ollama, LM Studio) ignore the Authorization header
+      // entirely. Send a non-empty placeholder since some servers reject empty
+      // Bearer values; never send the user's Codebuff key on this code path.
+      Authorization: `Bearer ${apiKey ?? 'unused'}`,
       'Content-Type': 'application/json',
       'user-agent': `ai-sdk/openai-compatible/${VERSION}/codebuff-custom-provider`,
     }),