nitro_intelligence 2.6.0 → 2.8.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 92e09bec20a682ff36923a40ad1dc9d3804c3bea3da199374859d5779034d524
-  data.tar.gz: 303cfc371638d3c14e41ad2610ca2de68944c5f0877db5ab77921ec77b192dd2
+  metadata.gz: 5f65b41b164172defbd342f01a3be82035ce33ce1f057a64700391fddacc8d51
+  data.tar.gz: 8c1086916735274f86154cebf70bc7cc91e30b2459da5e9adc3bb773e18f661d
 SHA512:
-  metadata.gz: 1d7b784fb71aa635ec0f6dc6388b6925312200e4cb757a6598b498ea0c8632bfd1413cbe6dd2a192244bd206d83f335a8865f3387d11d038092beeac65ae5397
-  data.tar.gz: 6225968bb8d513bffe785dbf91aba771bfdf0306571f8d3150489a839e8780c60b1ab7e15608f310ac56305c8dfd643b7b00a312485db3c3ee716d7573f895b7
+  metadata.gz: 9e8ed2298e83a75a257a91df4fd8ad1a64a86e53ef312f6c4b466bd5b9e88753c45af292fa1af3896f9fd8b7804222dbba54f23ed22962e28e4b7d20b398154e
+  data.tar.gz: 77f1fe28780cc5ea002754a102ae5b28046104597d8c18a76f2c5e4b50523df475f5e2ec2afd7d8159236621a435e1b2109e656bbd9e2c04cf8094feac8c7654

data/docs/README.md

@@ -18,10 +18,10 @@ NitroIntelligence.configure do |config|
 
   # Inference (LLM) settings
   config.inference_api_key  = "..."              # API key for the inference service
-  config.inference_base_url = "https://..."      # Base URL for the inference service
+  config.inference_base_url = "https://..."      # Base URL for the inference service (optional; defaults to the inference gateway)
 
   # Observability (Langfuse) settings
-  config.observability_base_url = "https://..."  # Base URL for the observability service
+  config.observability_base_url = "https://..."  # Base URL for the observability service (optional; defaults to Cerebro)
   config.observability_projects = [              # Array of project credential hashes
     {
       "slug"       => "my-feature-project",
@@ -83,10 +83,10 @@ end
 | `environment`            | `String`      | `"test"`              | Runtime environment name                                                                                                                                                                                   |
 | `cache_provider`         | cache store   | `NullCache`           | ActiveSupport-compatible cache store                                                                                                                                                                       |
 | `inference_api_key`      | `String`      | `""`                  | API key for the LLM inference service                                                                                                                                                                      |
-| `inference_base_url`     | `String`      | `""`                  | Base URL for the LLM inference service                                                                                                                                                                     |
-| `observability_base_url` | `String`      | `""`                  | Base URL for the Langfuse observability service                                                                                                                                                            |
+| `inference_base_url`     | `String`      | `"https://inference.powerhome.ai"` | Base URL for the LLM inference service. Defaults to the shared inference gateway, so only a host talking to a different one needs to set it                                                                 |
+| `observability_base_url` | `String`      | `"https://cerebro.powerhome.ai"` | Base URL for the Langfuse observability service. Defaults to Cerebro, so only a host talking to a different one needs to set it                                                                             |
 | `observability_projects` | `Array<Hash>` | `[]`                  | Langfuse project credentials (slug, id, public_key, secret_key)                                                                                                                                            |
-| `assistants_config`      | `Hash`        | `{}`                  | Assistants to make addressable by key. `base_url` (String) and `user_id` (String, default: `"default-user"`) are shared by every entry; `definitions` (Hash) holds one entry per assistant, keyed by what it is looked up with, each able to override a shared value. Without `definitions` the hash is read as credentials for a single `Assistants.new` — see [Assistants](#assistants) |
+| `assistants_config`      | `Hash`        | `{}`                  | Assistants to make addressable by key. `base_url` (String, default: `"https://assistants.powerhome.ai"`) and `user_id` (String, default: `"default-user"`) are shared by every entry; `definitions` (Hash) holds one entry per assistant, keyed by what it is looked up with, each able to override a shared value. Without `definitions` the hash is read as credentials for a single `Assistants.new`, which takes the same two defaults — see [Assistants](#assistants) |
 | `model_config`           | `Hash`        | `{}`                  | Model defaults and per-model settings. Top-level keys: `default_text_model`, `default_audio_transcription_model`, `default_image_model`, `default_text_to_speech_model`, and `models` (array of per-model hashes keyed by `name` and `type`, with type-specific options like `aspect_ratios`/`resolutions` for images or `voices`/`response_formats` for TTS) |
 
 ## Basic Usage
@@ -525,6 +525,7 @@ entry per assistant, keyed by what you look it up with:
 
 ```ruby
 config.assistants_config = {
+  # Optional. Defaults to the shared Assistants deployment.
   "base_url" => "https://nip-assistants.example.com",
   "user_id" => "my-app",
   "definitions" => {
@@ -562,9 +563,14 @@ convention, so nothing here is tied to one deployment's wiring.
 
 A key that resolves without them raises `Assistant::ConfigurationError`, naming every field it
 is missing at once so a host resolving them from elsewhere can see which lookup failed.
+`base_url` is not among them: an entry that says nothing about where to reach its assistant
+gets `https://assistants.powerhome.ai`, so only a host talking to a different deployment — a
+review environment, a local server — sets it.
 
 ### Without `definitions`
 
 `assistants_config` lacking `definitions` is read as keyword arguments for a single
 `Assistants` client, and `NitroIntelligence.assistants` returns that client rather than a
 registry. This is the shape that predates lookup by key; a host still on it is left alone.
+`base_url` is optional here too, and a configuration without one reaches the same shared
+deployment.

data/lib/nitro_intelligence/assistant.rb

@@ -1,3 +1,5 @@
+require "nitro_intelligence/assistants"
+
 module NitroIntelligence
   # One assistant resolved by name: the client for its deployment, plus the id every run has to
   # carry.
@@ -8,8 +10,10 @@ module NitroIntelligence
   class Assistant
     class ConfigurationError < StandardError; end
 
-    DEFAULT_USER_ID = "default-user".freeze
-    REQUIRED = %w[base_url api_key assistant_id].freeze
+    DEFAULT_USER_ID = Assistants::DEFAULT_USER_ID
+
+    # `base_url` is not among them: the client defaults it, so it cannot be missing.
+    REQUIRED = %w[api_key assistant_id].freeze
 
     # The key an assistant is filed and looked up under. Not its name: an entry usually carries
     # a `name` of its own, a human-readable label for the assistant's record, and the two are
@@ -25,7 +29,7 @@ module NitroIntelligence
     # application has no use for, such as the graph or the observability project it reports to.
     def initialize(key, base_url: nil, api_key: nil, assistant_id: nil, user_id: nil, **_kwargs)
       @key = key.to_s
-      @base_url = base_url.presence
+      @base_url = base_url.presence || Assistants::DEFAULT_BASE_URL
       @api_key = api_key.presence
       @assistant_id = assistant_id.presence
       @user_id = user_id.presence || DEFAULT_USER_ID
@@ -66,7 +70,7 @@ module NitroIntelligence
     # Reported together and named, since a host resolving these from somewhere else needs to
     # know which one it failed to supply.
     def validate!
-      values = { "base_url" => @base_url, "api_key" => @api_key, "assistant_id" => @assistant_id }
+      values = { "api_key" => @api_key, "assistant_id" => @assistant_id }
       missing = REQUIRED.select { |field| values[field].blank? }
       return if missing.empty?
 

data/lib/nitro_intelligence/assistant_registry.rb

@@ -59,7 +59,11 @@ module NitroIntelligence
               "Configured: #{keys.sort.join(', ').presence || '(none)'}"
       end
 
-      attributes = @config.slice(*SHARED_KEYS).merge(definition.to_h.deep_stringify_keys)
+      # Blank entry values are dropped rather than merged, so a definition built from a source
+      # that writes an unset field as null falls back to the shared setting instead of shadowing
+      # it with nothing.
+      entry = definition.to_h.deep_stringify_keys.compact_blank
+      attributes = @config.slice(*SHARED_KEYS).merge(entry)
       Assistant.new(key, **attributes.symbolize_keys)
     end
   end

data/lib/nitro_intelligence/assistants.rb

@@ -14,14 +14,20 @@ module NitroIntelligence
     # Assistants answers with a conflict when `ifExists: "raise"` is sent for a thread that already exists.
     THREAD_CONFLICT_CODE = 409
 
+    # Every assistant this gem is built for is served by the same deployment, so a client told
+    # nothing about where to reach one gets that deployment. A host talking to a different one
+    # -- a review environment, a local server -- still says so.
+    DEFAULT_BASE_URL = "https://assistants.powerhome.ai".freeze
+
+    DEFAULT_USER_ID = "default-user".freeze
+
     attr_reader :base_url, :user_id
 
-    def initialize(base_url:, api_key:, user_id: "default-user")
-      raise ConfigurationError, "base_url is required" if base_url.blank?
+    def initialize(api_key:, base_url: nil, user_id: DEFAULT_USER_ID)
       raise ConfigurationError, "api_key is required" if api_key.blank?
       raise ConfigurationError, "user_id is required" if user_id.blank?
 
-      @base_url = base_url
+      @base_url = base_url.presence || DEFAULT_BASE_URL
       @api_key = api_key
       @user_id = user_id
       @tool_call_review_validator = ToolCallReviewValidator.new
@@ -201,9 +207,19 @@ module NitroIntelligence
       raise RunError, run_response.body if run_response.code.to_i != 200
 
       run = JSON.parse(run_response.body)
+      raise_run_error!(run, RunError)
+
       Array(run["messages"]).last&.dig("content")
     end
 
+    def raise_run_error!(run, error)
+      failure = run["__error__"]
+      return if failure.blank?
+
+      detail = failure.is_a?(Hash) ? [failure["error"], failure["message"]].compact.join(": ") : failure.to_s
+      raise error, detail
+    end
+
     def resume_run(thread_id:, assistant_id:, resume:, context:)
       run_response = post(
         path: "/threads/#{thread_id}/runs/wait",
@@ -218,7 +234,10 @@ module NitroIntelligence
 
       raise ThreadResumptionError, run_response.body if run_response.code.to_i != 200
 
-      JSON.parse(run_response.body)
+      run = JSON.parse(run_response.body)
+      raise_run_error!(run, ThreadResumptionError)
+
+      run
     end
 
     def interrupted?(thread)

data/lib/nitro_intelligence/client/handlers/base_handler.rb

@@ -61,10 +61,9 @@ module NitroIntelligence
 
       private
 
-        # `last_response` carries the HTTP metadata of the response a typed model was
-        # built from. The client leaves it unset on nested and locally constructed
-        # models, and on endpoints returning raw or binary payloads, so both the
-        # method and its value are optional.
+        # `last_response` carries the HTTP metadata of the response a model was built
+        # from. The client leaves it unset on nested and locally constructed models,
+        # so both the method and its value are optional.
         def response_headers(response)
           return nil unless response.respond_to?(:last_response)
 

data/lib/nitro_intelligence/client/handlers/observed/text_to_speech_handler.rb

@@ -79,11 +79,13 @@ module NitroIntelligence
               output = handle_text_to_speech_upload(tempfile, trace_id)
             end
 
-            # We only get StringIO object as a response, so there are no usage details
-            # and no resolved model to record. The requested model and the input are
-            # already on the observation from before the request ran.
+            # Usage details and the resolved model are read off a response body, and
+            # this endpoint's body is audio, so neither is available here. The
+            # requested model and the input are already on the observation from
+            # before the request ran.
             trace_attributes = {
               output:,
+              cost_details: @base_handler.cost_details(tts),
             }
 
             [tts, trace_attributes]

data/lib/nitro_intelligence/configuration.rb

@@ -12,9 +12,9 @@ module NitroIntelligence
     config_accessor :environment, default: "test"
     config_accessor :assistants_config, default: {}
     config_accessor :inference_api_key, default: ""
-    config_accessor :inference_base_url, default: ""
+    config_accessor :inference_base_url, default: "https://inference.powerhome.ai"
     config_accessor :model_config, default: {}
-    config_accessor :observability_base_url, default: ""
+    config_accessor :observability_base_url, default: "https://cerebro.powerhome.ai"
     config_accessor :observability_projects, default: []
     config_accessor :observability_user_id, default: ""
 

data/lib/nitro_intelligence/version.rb

@@ -1,3 +1,3 @@
 module NitroIntelligence
-  VERSION = "2.6.0".freeze
+  VERSION = "2.8.0".freeze
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nitro_intelligence
 version: !ruby/object:Gem::Version
-  version: 2.6.0
+  version: 2.8.0
 platform: ruby
 authors:
 - Igor Artemenko
@@ -57,14 +57,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.79'
+        version: '0.86'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.79'
+        version: '0.86'
 description: The Ruby client for Nitro Intelligence
 email:
 - igor.artemenko@powerhrg.com