chore: run ruff format across nodes/src/nodes/ for consistent baseline

nodes/src/nodes/accessibility_describe/accessibility_vision.py

@@ -67,7 +67,7 @@
 
 # Spatial format prompt modifiers
 SPATIAL_PROMPTS = {
-    'clock': '\n\nUse clock positions for spatial references (12 o\'clock = straight ahead).',
+    'clock': "\n\nUse clock positions for spatial references (12 o'clock = straight ahead).",
     'relative': '\n\nUse relative directions (left, right, ahead, behind) for spatial references.',
     'both': '\n\nUse both clock positions and relative directions for spatial references.',
 }
@@ -98,31 +98,18 @@ def __init__(self, provider: str, connConfig: dict[str, Any], bag: dict[str, Any
         spatial_format = config.get('accessibility.spatialFormat', 'clock')
 
         # Build system prompt with config modifiers
-        self._system_prompt = (
-            config.get('accessibility.systemPrompt')
-            or config.get('systemPrompt')
-            or DEFAULT_SYSTEM_PROMPT
-        )
+        self._system_prompt = config.get('accessibility.systemPrompt') or config.get('systemPrompt') or DEFAULT_SYSTEM_PROMPT
         self._system_prompt += HAZARD_PROMPTS.get(hazard_priority, '')
         self._system_prompt += SPATIAL_PROMPTS.get(spatial_format, '')
 
-        self._prompt = (
-            config.get('accessibility.prompt')
-            or config.get('prompt')
-            or DEFAULT_PROMPT
-        )
+        self._prompt = config.get('accessibility.prompt') or config.get('prompt') or DEFAULT_PROMPT
 
         if not api_key:
-            raise ValueError(
-                'Missing Google AI API key. Get one at https://aistudio.google.com/apikey'
-            )
+            raise ValueError('Missing Google AI API key. Get one at https://aistudio.google.com/apikey')
 
         # Validate the API key format
         if api_key.startswith('sk-'):
-            raise ValueError(
-                'Invalid API key format. This appears to be an OpenAI key. '
-                'Please provide a Google AI API key.'
-            )
+            raise ValueError('Invalid API key format. This appears to be an OpenAI key. Please provide a Google AI API key.')
 
         try:
             self._client = genai.Client(api_key=api_key)
@@ -157,7 +144,7 @@ def _format_user_error(self, error_msg: str) -> str:
         if any(phrase in error_lower for phrase in ['invalid input', 'bad request', '400']):
             return 'Invalid input. Please check your image format and prompt.'
         if any(phrase in error_lower for phrase in ['model not found', 'unavailable', 'not supported']):
-            return f'Model \'{self._model}\' is currently unavailable. Please try a different model.'
+            return f"Model '{self._model}' is currently unavailable. Please try a different model."
         if any(phrase in error_lower for phrase in ['timeout', 'timed out']):
             return 'Request timed out. Please try again.'
         if any(phrase in error_lower for phrase in ['content policy', 'safety', 'blocked']):
@@ -199,9 +186,7 @@ def chat(self, question: Question) -> Answer:
             mime_type = header.split(':')[1].split(';')[0]
             image_bytes = base64.b64decode(b64_data)
         except (ValueError, IndexError, base64.binascii.Error) as e:
-            raise ValueError(
-                'Malformed image data URL. Expected format: data:<mime>;base64,<data>'
-            ) from e
+            raise ValueError('Malformed image data URL. Expected format: data:<mime>;base64,<data>') from e
 
         # Build request contents once (deterministic, no need to rebuild per retry)
         contents = [
@@ -233,7 +218,7 @@ def chat(self, question: Question) -> Answer:
             except Exception as e:
                 last_error = e
                 if attempt < max_retries and self._shouldRetry(e):
-                    delay = base_delay * (2 ** attempt)
+                    delay = base_delay * (2**attempt)
                     time.sleep(delay)
                     continue
                 break

nodes/src/nodes/agent_crewai/crewai.py

@@ -198,12 +198,7 @@ def _invoke_tool(tool_name: str, input: Any = None, kwargs: Optional[Dict[str, A
         agent_obj = Agent(
             role='Assistant',
             goal='Solve the user request using available tools when helpful.',
-            backstory=(
-                'You are an agent node in a tool-invocation hierarchy. '
-                'You may call tools wired to you via the host tools interface. '
-                'When a tool is needed, call it; otherwise respond directly. '
-                'Follow any additional instructions exactly.'
-            ),
+            backstory=('You are an agent node in a tool-invocation hierarchy. You may call tools wired to you via the host tools interface. When a tool is needed, call it; otherwise respond directly. Follow any additional instructions exactly.'),
             tools=tools_for_agent,
             llm=llm,
             verbose=False,

nodes/src/nodes/agent_langchain/IInstance.py

@@ -46,4 +46,3 @@ def invoke(self, param: Any) -> Any:  # noqa: ANN401
         if isinstance(op, str) and op.startswith('tool.'):
             return self.IGlobal.agent.handle_invoke(self, param)
         return super().invoke(param)
-

nodes/src/nodes/agent_langchain/langchain.py

@@ -67,7 +67,7 @@ def _llm_type(self) -> str:
             def _identifying_params(self) -> Dict[str, Any]:
                 return {'framework': 'rocketride', 'adapter': 'tool_calling_json'}
 
-            def bind_tools(self, tools: Any, **kwargs: Any) -> "RocketRideToolCallingChatModel":
+            def bind_tools(self, tools: Any, **kwargs: Any) -> 'RocketRideToolCallingChatModel':
                 try:
                     self._bound_tools = _normalize_bound_tools(tools)
                 except Exception:
@@ -448,4 +448,3 @@ def _safe_str(v: Any) -> str:
         return '' if v is None else str(v)
     except Exception:
         return ''
-

nodes/src/nodes/agent_rocketride/IGlobal.py

@@ -42,6 +42,7 @@ def beginGlobal(self) -> None:
         config handling is needed here.
         """
         from .rocketride_agent import RocketRideDriver
+
         self.agent = RocketRideDriver(self)
 
     def endGlobal(self) -> None:

nodes/src/nodes/agent_rocketride/executor.py

@@ -81,6 +81,7 @@
 # Structural summary (_describe)
 # ---------------------------------------------------------------------------
 
+
 def _describe(value: Any, depth: int = 0) -> str:
     """Return a compact structural summary of *value* for LLM context.
 
@@ -118,9 +119,7 @@ def _describe(value: Any, depth: int = 0) -> str:
         if isinstance(first, dict):
             # Collect field names from up to 5 rows to handle sparse rows
             # where early rows may be missing fields that appear later.
-            keys = list(dict.fromkeys(
-                k for row in value[:5] if isinstance(row, dict) for k in row
-            ))
+            keys = list(dict.fromkeys(k for row in value[:5] if isinstance(row, dict) for k in row))
             lines = [f'{n} items, fields: {keys}']
             # Show first 2 rows as sample data so the LLM can see real values
             for i, row in enumerate(value[:2]):
@@ -152,6 +151,7 @@ def _describe_dict(d: dict, depth: int) -> str:
 # Template resolution
 # ---------------------------------------------------------------------------
 
+
 def _memory_get(key: str, host: AgentHost) -> Any:
     """Fetch a raw value from the memory store.
 
@@ -298,6 +298,7 @@ def resolve_answer_refs(answer: str, host: AgentHost) -> str:
 # Wave result storage
 # ---------------------------------------------------------------------------
 
+
 def _auto_key(wave_name: str, idx: int) -> str:
     """Generate a memory key scoped to a wave and call index.
 
@@ -335,6 +336,7 @@ def _store_and_preview(tool: str, key: str, result: Any, host: AgentHost) -> Dic
 # Wave executor
 # ---------------------------------------------------------------------------
 
+
 def _execute_wave_calls(
     wave: List[Dict[str, Any]],
     *,
@@ -356,10 +358,7 @@ def _execute_wave_calls(
 
     # Tag each call with its auto-generated memory key before parallelism so
     # the key assignment is deterministic and order-preserving.
-    tagged: List[Dict[str, Any]] = [
-        {**call, '_key': _auto_key(wave_name, i)}
-        for i, call in enumerate(wave)
-    ]
+    tagged: List[Dict[str, Any]] = [{**call, '_key': _auto_key(wave_name, i)} for i, call in enumerate(wave)]
 
     def _run_one(call: Dict[str, Any]) -> Dict[str, Any]:
         """Execute a single tool call and return a result dict."""
@@ -411,8 +410,11 @@ def _run_one(call: Dict[str, Any]) -> Dict[str, Any]:
                         total_items = len(value)
                         preview = json.dumps(value[:_PEEK_MAX_ARRAY_ITEMS], ensure_ascii=False)
                         return {
-                            'tool': tool, 'key': key, 'path': path,
-                            'preview': preview, 'truncated': True,
+                            'tool': tool,
+                            'key': key,
+                            'path': path,
+                            'preview': preview,
+                            'truncated': True,
                             'returned_items': _PEEK_MAX_ARRAY_ITEMS,
                             'total_items': total_items,
                         }
@@ -426,10 +428,14 @@ def _run_one(call: Dict[str, Any]) -> Dict[str, Any]:
                     value = json.dumps(value, ensure_ascii=False, indent=2)
                 offset = int(args.get('offset', 0))
                 length = int(args.get('length', _PEEK_DEFAULT_LENGTH))
-                chunk = value[offset:offset + length]
+                chunk = value[offset : offset + length]
                 return {
-                    'tool': tool, 'key': key, 'preview': chunk,
-                    'offset': offset, 'length': len(chunk), 'total_chars': len(value),
+                    'tool': tool,
+                    'key': key,
+                    'preview': chunk,
+                    'offset': offset,
+                    'length': len(chunk),
+                    'total_chars': len(value),
                 }
 
             # Regular tool — route through the host's tool pipeline which

nodes/src/nodes/agent_rocketride/rocketride_agent.py

@@ -46,6 +46,7 @@
 # Can be overridden via the ``max_waves`` node configuration field.
 _DEFAULT_MAX_WAVES = 10
 
+
 class RocketRideDriver(AgentBase):
     """
     RocketRide Wave framework driver.

nodes/src/nodes/anonymize/IInstance.py

@@ -27,15 +27,10 @@
 
 class IInstance(IInstanceBase):
     IGlobal: IGlobal  # Reference to a global context providing recognizer functionality
-    
+
     # Default PII labels for zero-shot NER when no classifications provided
-    DEFAULT_PII_LABELS = [
-        'person', 'name', 'email', 'phone number', 'address',
-        'social security number', 'credit card number', 'date of birth',
-        'organization', 'company', 'location', 'ip address',
-        'bank account', 'passport number', 'driver license'
-    ]
-
+    DEFAULT_PII_LABELS = ['person', 'name', 'email', 'phone number', 'address', 'social security number', 'credit card number', 'date of birth', 'organization', 'company', 'location', 'ip address', 'bank account', 'passport number', 'driver license']
+
     #
     # Current object context properties
     #
@@ -59,7 +54,7 @@ def closing(self):
         if not self.has_classifications:
             # No classifications received - anonymize with default PII labels
             self.target_object_text = self.IGlobal.recognizer.process(self.target_object_text, self.DEFAULT_PII_LABELS)
-        
+
         # Resume the writeText lane
         self.instance.writeText(self.target_object_text)
 

nodes/src/nodes/anonymize/anonymize.py

@@ -21,6 +21,7 @@
 # SOFTWARE.
 # =============================================================================
 
+
 def anonymize(text: str, matches, anonymize_char: str = '*') -> str:
     """Replace specified segments with a sequence of anonymization characters.
 

nodes/src/nodes/anonymize/glinerRecognizer.py

@@ -36,7 +36,7 @@ class GliNERRecognizer:
     def __init__(self, provider: str, connConfig: Dict[str, Any], bag: Dict[str, Any]):
         """
         Initialize the GLiNER Recognizer.
-        
+
         Uses ai.common.models.GLiNER which automatically routes to model server
         if --modelserver flag is present, otherwise runs locally.
         """
@@ -50,7 +50,7 @@ def __init__(self, provider: str, connConfig: Dict[str, Any], bag: Dict[str, Any
         enginePath = expand('%execPath%')
         rule_file_path = os.path.join(enginePath, 'nucleuz', 'rulePack.dat')
         self.ruleParser = RuleParser(rule_file_path)
-        
+
         # Use ai.common.models.GLiNER - auto-detects local vs model server mode
         self.model = GLiNER(self.model_name)
 
@@ -182,34 +182,34 @@ def process_chunk(chunk_idx):
     def process(self, text: str, labels: list, existing_matches: list = None) -> str:
         """
         Core anonymization method - detects entities using GLiNER and masks them.
-        
+
         Args:
             text: The text to anonymize
             labels: Entity labels to detect
             existing_matches: Optional list of (offset, length) tuples from classifications
-            
+
         Returns:
             Anonymized text with detected entities replaced by anonymize_char
         """
         if not text:
             return text
-        
+
         # Run NER prediction
         ner_results = self.predict(text, labels)
         ner_matches = self.convert_ner_results_to_matches(ner_results)
-        
+
         debug(f'Anonymize: Detected {len(ner_results)} entities')
-        
+
         # Combine with existing matches (from classifications)
         all_matches = list(existing_matches or []) + ner_matches
-        
+
         if not all_matches:
             debug('Anonymize: No entities to mask')
             return text
-        
+
         # Sort by offset and apply masking
         all_matches_sorted = sorted(all_matches, key=lambda x: x[0])
-        
+
         return _anonymize(text, all_matches_sorted, self.anonymize_char)
 
     def handleClassifications(self, classifications: dict, target_object_text: str, classificationPolicy: any, classificationRules: any):
@@ -234,9 +234,6 @@ def handleClassifications(self, classifications: dict, target_object_text: str,
         labels = self.ruleParser.get_rules_names(unique_id_refs) + rules
 
         # Extract existing matches from classifications (offset, length tuples)
-        existing_matches = list(
-            (m['offset'], m['length']) 
-            for m in ((m.get('location', {}).get('inChars') or m) for m in text_matches)
-        )
+        existing_matches = list((m['offset'], m['length']) for m in ((m.get('location', {}).get('inChars') or m) for m in text_matches))
 
         return self.process(target_object_text, labels, existing_matches)

nodes/src/nodes/audio_transcribe/IGlobal.py

@@ -66,10 +66,7 @@ def transcribe(self, audio: Any) -> List[SimpleNamespace]:
             )
 
         segments = result.get('$segments') or []
-        return [
-            SimpleNamespace(text=s.get('text', ''), start=s.get('start', 0.0), end=s.get('end', 0.0))
-            for s in segments
-        ]
+        return [SimpleNamespace(text=s.get('text', ''), start=s.get('start', 0.0), end=s.get('end', 0.0)) for s in segments]
 
     def _audio_to_pcm_bytes(self, audio: Any) -> bytes:
         """Convert audio (bytes or float32 numpy) to PCM int16 bytes (16 kHz mono)."""

nodes/src/nodes/autopipe/IGlobal.py

@@ -168,7 +168,6 @@ def isSet(key: str):
                 providerStore, configStore = Config.getMultiProviderConfig('store', autopipeConfig)
                 pushLocal(getFilter('vector_1', providerStore, configStore))
 
-
             pass
 
         # Now, add all the filters we figured out

nodes/src/nodes/chart_chartjs/IInstance.py

@@ -70,9 +70,7 @@ def invoke(self, param: Any) -> Any:  # noqa: ANN401
         if driver._llm_invoke is None:
             llm_invoker = _make_llm_invoker(self.instance)
             if llm_invoker is None:
-                raise RuntimeError(
-                    'Chart generator requires an LLM node connected to the pipeline.'
-                )
+                raise RuntimeError('Chart generator requires an LLM node connected to the pipeline.')
             driver.set_llm_invoker(llm_invoker)
 
         return driver.handle_invoke(param)