agenta 0.57.0__py3-none-any.whl → 0.63.2__py3-none-any.whl

agenta/sdk/workflows/sandbox.py

@@ -0,0 +1,118 @@
+from typing import Union, Text, Dict, Any
+
+from RestrictedPython import safe_builtins, compile_restricted, utility_builtins
+from RestrictedPython.Eval import (
+    default_guarded_getiter,
+    default_guarded_getitem,
+)
+from RestrictedPython.Guards import (
+    guarded_iter_unpack_sequence,
+    full_write_guard,
+)
+
+
+def is_import_safe(python_code: Text) -> bool:
+    """Checks if the imports in the python code contains a system-level import.
+
+    Args:
+        python_code (str): The Python code to be executed
+
+    Returns:
+        bool - module is secured or not
+    """
+
+    disallowed_imports = ["os", "subprocess", "threading", "multiprocessing"]
+    for import_ in disallowed_imports:
+        if import_ in python_code:
+            return False
+    return True
+
+
+def execute_code_safely(
+    app_params: Dict[str, Any],
+    inputs: Dict[str, Any],
+    output: Union[dict, str],
+    correct_answer: Any,  # for backward compatibility reasons
+    code: Text,
+) -> Union[float, None]:
+    """
+    Execute the provided Python code safely using RestrictedPython.
+
+    Args:
+        - app_params (Dict[str, str]): The parameters of the app variant.
+        - inputs (dict): Inputs to be used during code execution.
+        - output (str): The output of the app variant after being called.
+        - correct_answer (str): The correct answer (or target) of the app variant.
+        - code (Text): The Python code to be executed.
+        - datapoint (Dict[str, str]): The test datapoint.
+
+    Returns:
+    - (float): Result of the execution if successful. Should be between 0 and 1.
+    - None if execution fails or result is not a float between 0 and 1.
+    """
+    # Define the available built-ins
+    local_builtins = safe_builtins.copy()
+
+    # Add the __import__ built-in function to the local builtins
+    local_builtins["__import__"] = __import__
+
+    # Define supported packages
+    allowed_imports = [
+        "math",
+        "random",
+        "datetime",
+        "json",
+        "requests",
+        "typing",
+    ]
+
+    # Create a dictionary to simulate allowed imports
+    allowed_modules = {}
+    for package_name in allowed_imports:
+        allowed_modules[package_name] = __import__(package_name)
+
+    # Add the allowed modules to the local built-ins
+    local_builtins.update(allowed_modules)
+    local_builtins.update(utility_builtins)
+
+    # Define the environment for the code execution
+    environment = {
+        "_getiter_": default_guarded_getiter,
+        "_getitem_": default_guarded_getitem,
+        "_iter_unpack_sequence_": guarded_iter_unpack_sequence,
+        "_write_": full_write_guard,
+        "__builtins__": local_builtins,
+    }
+
+    # Compile the code in a restricted environment
+    byte_code = compile_restricted(code, filename="<inline>", mode="exec")
+
+    # Call the evaluation function, extract the result if it exists
+    # and is a float between 0 and 1
+    try:
+        # Execute the code
+        exec(byte_code, environment)
+
+        # Call the evaluation function, extract the result
+        result = environment["evaluate"](app_params, inputs, output, correct_answer)
+
+        # Attempt to convert result to float
+        if isinstance(result, (float, int, str)):
+            try:
+                result = float(result)
+            except ValueError as e:
+                raise ValueError(f"Result cannot be converted to float: {e}")
+
+        if not isinstance(result, float):
+            raise TypeError(f"Result is not a float after conversion: {type(result)}")
+
+        return result
+
+    except KeyError as e:
+        raise KeyError(f"Missing expected key in environment: {e}")
+
+    except SyntaxError as e:
+        raise SyntaxError(f"Syntax error in provided code: {e}")
+
+    except Exception as e:
+        raise RuntimeError(f"Error during code execution: {e}")

agenta/sdk/workflows/utils.py

@@ -1,17 +1,314 @@
-from typing import Optional, Tuple
+# /agenta/sdk/workflows/utils.py
 
+from typing import Optional, Tuple, Callable
 
-async def parse_service_uri(
+from agenta.sdk.models.workflows import WorkflowServiceInterface
+
+from agenta.sdk.workflows.handlers import (
+    echo_v0,
+    auto_exact_match_v0,
+    auto_regex_test_v0,
+    field_match_test_v0,
+    auto_webhook_test_v0,
+    auto_custom_code_run_v0,
+    auto_ai_critique_v0,
+    auto_starts_with_v0,
+    auto_ends_with_v0,
+    auto_contains_v0,
+    auto_contains_any_v0,
+    auto_contains_all_v0,
+    auto_contains_json_v0,
+    auto_json_diff_v0,
+    auto_levenshtein_distance_v0,
+    auto_similarity_match_v0,
+    auto_semantic_similarity_v0,
+    completion_v0,
+    chat_v0,
+)
+
+from agenta.sdk.workflows.interfaces import (
+    echo_v0_interface,
+    auto_exact_match_v0_interface,
+    auto_regex_test_v0_interface,
+    field_match_test_v0_interface,
+    auto_webhook_test_v0_interface,
+    auto_custom_code_run_v0_interface,
+    auto_ai_critique_v0_interface,
+    auto_starts_with_v0_interface,
+    auto_ends_with_v0_interface,
+    auto_contains_v0_interface,
+    auto_contains_any_v0_interface,
+    auto_contains_all_v0_interface,
+    auto_contains_json_v0_interface,
+    auto_json_diff_v0_interface,
+    auto_levenshtein_distance_v0_interface,
+    auto_similarity_match_v0_interface,
+    auto_semantic_similarity_v0_interface,
+    completion_v0_interface,
+    chat_v0_interface,
+)
+
+
+from agenta.sdk.workflows.configurations import (
+    echo_v0_configuration,
+    auto_exact_match_v0_configuration,
+    auto_regex_test_v0_configuration,
+    field_match_test_v0_configuration,
+    auto_webhook_test_v0_configuration,
+    auto_custom_code_run_v0_configuration,
+    auto_ai_critique_v0_configuration,
+    auto_starts_with_v0_configuration,
+    auto_ends_with_v0_configuration,
+    auto_contains_v0_configuration,
+    auto_contains_any_v0_configuration,
+    auto_contains_all_v0_configuration,
+    auto_contains_json_v0_configuration,
+    auto_json_diff_v0_configuration,
+    auto_levenshtein_distance_v0_configuration,
+    auto_similarity_match_v0_configuration,
+    auto_semantic_similarity_v0_configuration,
+    completion_v0_configuration,
+    chat_v0_configuration,
+)
+
+INTERFACE_REGISTRY: dict = dict(
+    agenta={
+        "built-in": dict(
+            echo=dict(v0=echo_v0_interface),
+            auto_exact_match=dict(v0=auto_exact_match_v0_interface),
+            auto_regex_test=dict(v0=auto_regex_test_v0_interface),
+            field_match_test=dict(v0=field_match_test_v0_interface),
+            auto_webhook_test=dict(v0=auto_webhook_test_v0_interface),
+            auto_custom_code_run=dict(v0=auto_custom_code_run_v0_interface),
+            auto_ai_critique=dict(v0=auto_ai_critique_v0_interface),
+            auto_starts_with=dict(v0=auto_starts_with_v0_interface),
+            auto_ends_with=dict(v0=auto_ends_with_v0_interface),
+            auto_contains=dict(v0=auto_contains_v0_interface),
+            auto_contains_any=dict(v0=auto_contains_any_v0_interface),
+            auto_contains_all=dict(v0=auto_contains_all_v0_interface),
+            auto_contains_json=dict(v0=auto_contains_json_v0_interface),
+            auto_json_diff=dict(v0=auto_json_diff_v0_interface),
+            auto_levenshtein_distance=dict(v0=auto_levenshtein_distance_v0_interface),
+            auto_similarity_match=dict(v0=auto_similarity_match_v0_interface),
+            auto_semantic_similarity=dict(v0=auto_semantic_similarity_v0_interface),
+            completion=dict(v0=completion_v0_interface),
+            chat=dict(v0=chat_v0_interface),
+        ),
+    },
+)
+
+CONFIGURATION_REGISTRY: dict = dict(
+    agenta={
+        "built-in": dict(
+            echo=dict(v0=echo_v0_configuration),
+            auto_exact_match=dict(v0=auto_exact_match_v0_configuration),
+            auto_regex_test=dict(v0=auto_regex_test_v0_configuration),
+            field_match_test=dict(v0=field_match_test_v0_configuration),
+            auto_webhook_test=dict(v0=auto_webhook_test_v0_configuration),
+            auto_custom_code_run=dict(v0=auto_custom_code_run_v0_configuration),
+            auto_ai_critique=dict(v0=auto_ai_critique_v0_configuration),
+            auto_starts_with=dict(v0=auto_starts_with_v0_configuration),
+            auto_ends_with=dict(v0=auto_ends_with_v0_configuration),
+            auto_contains=dict(v0=auto_contains_v0_configuration),
+            auto_contains_any=dict(v0=auto_contains_any_v0_configuration),
+            auto_contains_all=dict(v0=auto_contains_all_v0_configuration),
+            auto_contains_json=dict(v0=auto_contains_json_v0_configuration),
+            auto_json_diff=dict(v0=auto_json_diff_v0_configuration),
+            auto_levenshtein_distance=dict(
+                v0=auto_levenshtein_distance_v0_configuration
+            ),
+            auto_similarity_match=dict(v0=auto_similarity_match_v0_configuration),
+            auto_semantic_similarity=dict(v0=auto_semantic_similarity_v0_configuration),
+            completion=dict(v0=completion_v0_configuration),
+            chat=dict(v0=chat_v0_configuration),
+        ),
+    },
+)
+
+# Global registry for workflow handlers organized by URI structure.
+#
+# URI Format: provider:kind:key:version
+#
+# Structure:
+#   HANDLER_REGISTRY[provider][kind][key][version] = handler_callable
+#
+# Components:
+#   - provider: The source/namespace of the handler (e.g., "agenta", "user")
+#   - kind: The category/type of handler (e.g., "built-in", "custom")
+#   - key: The unique identifier for the handler (e.g., "echo", "auto_exact_match", "module.function_name")
+#   - version: The version identifier (e.g., "v0", "v1", "latest")
+#
+# Examples:
+#   - URI: "agenta:built-in:echo:v0"
+#     Access: HANDLER_REGISTRY["agenta"]["built-in"]["echo"]["v0"]
+#
+#   - URI: "user:custom:mymodule.my_workflow:latest"
+#     Access: HANDLER_REGISTRY["user"]["custom"]["mymodule.my_workflow"]["latest"]
+#
+# Usage:
+#   - register_handler(fn, uri) - Registers a new handler with the given URI
+#   - retrieve_handler(uri) - Retrieves a handler by its URI
+#   - retrieve_interface(uri) - Retrieves the interface configuration for a handler
+#   - retrieve_configuration(uri) - Retrieves default parameters for a handler
+#
+# The registry supports automatic URI generation for user-defined workflows:
+#   If no URI is provided, register_handler() generates: "user:custom:{module}.{name}:latest"
+HANDLER_REGISTRY: dict = dict(
+    agenta={
+        "built-in": dict(
+            echo=dict(v0=echo_v0),
+            auto_exact_match=dict(v0=auto_exact_match_v0),
+            auto_regex_test=dict(v0=auto_regex_test_v0),
+            field_match_test=dict(v0=field_match_test_v0),
+            auto_webhook_test=dict(v0=auto_webhook_test_v0),
+            auto_custom_code_run=dict(v0=auto_custom_code_run_v0),
+            auto_ai_critique=dict(v0=auto_ai_critique_v0),
+            auto_starts_with=dict(v0=auto_starts_with_v0),
+            auto_ends_with=dict(v0=auto_ends_with_v0),
+            auto_contains=dict(v0=auto_contains_v0),
+            auto_contains_any=dict(v0=auto_contains_any_v0),
+            auto_contains_all=dict(v0=auto_contains_all_v0),
+            auto_contains_json=dict(v0=auto_contains_json_v0),
+            auto_json_diff=dict(v0=auto_json_diff_v0),
+            auto_levenshtein_distance=dict(v0=auto_levenshtein_distance_v0),
+            auto_similarity_match=dict(v0=auto_similarity_match_v0),
+            auto_semantic_similarity=dict(v0=auto_semantic_similarity_v0),
+            completion=dict(v0=completion_v0),
+            chat=dict(v0=chat_v0),
+        ),
+    },
+)
+
+
+def parse_uri(
     uri: str,
 ) -> Tuple[Optional[str], Optional[str], Optional[str], Optional[str]]:
     if not uri or not uri.strip():
         return None, None, None, None
 
-    # uri ~ [<provider>|empty]:<kind>:<key>:[<version>|'latest'|empty]
-
     parts = uri.split(":")
 
-    if len(parts) != 4:
+    # 1 → key
+    # 2 → kind:key
+    # 3 → provider:kind:key
+    # 4 → provider:kind:key:version
+    if len(parts) == 1:
+        provider, kind, key, version = "agenta", "built-in", parts[0], "latest"
+    elif len(parts) == 2:
+        provider, kind, key, version = "agenta", parts[0], parts[1], "latest"
+    elif len(parts) == 3:
+        provider, kind, key, version = parts[0], parts[1], parts[2], "latest"
+    elif len(parts) == 4:
+        provider, kind, key, version = parts[0], parts[1], parts[2], parts[3]
+    else:
         return None, None, None, None
 
-    return tuple(parts)
+    return provider, kind, key, version
+
+
+def register_handler(fn: Callable, uri: Optional[str] = None) -> str:
+    """Register a handler function in the global handler registry.
+
+    Stores a callable in the HANDLER_REGISTRY with a hierarchical URI structure
+    of provider:kind:key:version. If no URI is provided, generates one automatically
+    using the function's module and name (user:custom:module.name:latest).
+
+    The URI is parsed into components and used to create nested dictionary entries
+    in the registry for later retrieval by retrieve_handler().
+
+    Args:
+        fn: The callable function to register
+        uri: Optional URI string in format "provider:kind:key:version".
+             If None, auto-generates "user:custom:{module}.{name}:latest"
+
+    Returns:
+        The URI string used for registration
+
+    Raises:
+        ValueError: If the URI is invalid or missing required components
+
+    Example:
+        >>> def my_workflow(): pass
+        >>> uri = register_handler(my_workflow, "user:custom:my_workflow:v1")
+        >>> uri
+        'user:custom:my_workflow:v1'
+    """
+    if not uri:
+        key = f"{fn.__module__}.{fn.__name__}"
+        uri = f"user:custom:{key}:latest"
+
+    provider, kind, key, version = parse_uri(uri)  # type: ignore
+
+    if not provider or not kind or not key or not version:
+        raise ValueError(f"Invalid URI: {uri}")
+
+    HANDLER_REGISTRY.setdefault(provider, {}).setdefault(kind, {}).setdefault(
+        key, {}
+    ).setdefault(version, fn)
+
+    return uri
+
+
+def _get_with_latest(
+    registry: dict,
+    provider: Optional[str] = None,
+    kind: Optional[str] = None,
+    key: Optional[str] = None,
+    version: Optional[str] = None,
+):
+    kind_dict = registry.get(provider, {}).get(kind, {}).get(key, {})
+
+    if not isinstance(kind_dict, dict) or not kind_dict:
+        return None
+
+    if version == "latest":
+        # if "latest" explicitly exists, prefer it
+        if "latest" in kind_dict:
+            return kind_dict.get("latest")
+
+        # collect keys of the form vN
+        candidates = [
+            (int(v[1:]), v)
+            for v in kind_dict.keys()
+            if isinstance(v, str) and v.startswith("v") and v[1:].isdigit()
+        ]
+        if not candidates:
+            return None
+        # get the highest int N
+        _, best_key = max(candidates, key=lambda x: x[0])
+        return kind_dict.get(best_key)
+
+    return kind_dict.get(version)
+
+
+def retrieve_handler(uri: Optional[str] = None) -> Optional[Callable]:
+    if not uri:
+        return None
+    provider, kind, key, version = parse_uri(uri)
+
+    return _get_with_latest(HANDLER_REGISTRY, provider, kind, key, version)
+
+
+def retrieve_interface(uri: Optional[str] = None) -> Optional[WorkflowServiceInterface]:
+    if not uri:
+        return None
+    provider, kind, key, version = parse_uri(uri)
+
+    return _get_with_latest(INTERFACE_REGISTRY, provider, kind, key, version)
+
+
+def retrieve_configuration(uri: Optional[str] = None) -> Optional[dict]:
+    if not uri:
+        return None
+    provider, kind, key, version = parse_uri(uri)
+
+    return _get_with_latest(CONFIGURATION_REGISTRY, provider, kind, key, version)
+
+
+def is_custom_uri(uri: Optional[str] = None) -> bool:
+    if not uri:
+        return True
+
+    provider, kind, key, version = parse_uri(uri)
+
+    return provider == "user" and kind == "custom"

{agenta-0.57.0.dist-info → agenta-0.63.2.dist-info}/METADATA

@@ -1,42 +1,44 @@
 Metadata-Version: 2.4
 Name: agenta
-Version: 0.57.0
+Version: 0.63.2
 Summary: The SDK for agenta is an open-source LLMOps platform.
 Keywords: LLMOps,LLM,evaluation,prompt engineering
 Author: Mahmoud Mabrouk
 Author-email: mahmoud@agenta.ai
-Requires-Python: >=3.9,<4.0
+Requires-Python: >=3.11,<4.0
 Classifier: Intended Audience :: Developers
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: Python :: 3.9
 Classifier: Topic :: Software Development :: Libraries
 Requires-Dist: decorator (>=5.2.1,<6.0.0)
 Requires-Dist: fastapi (>=0.116.0,<0.117.0)
+Requires-Dist: google-auth (>=2.23,<3)
 Requires-Dist: h11 (>=0.16.0,<0.17.0)
 Requires-Dist: httpx (>=0.28.0,<0.29.0)
 Requires-Dist: huggingface-hub (<0.31.0)
 Requires-Dist: importlib-metadata (>=8.0.0,<9.0)
 Requires-Dist: jinja2 (>=3.1.6,<4.0.0)
-Requires-Dist: litellm (==1.76.0)
-Requires-Dist: openai (>=1.106.0,<2.0.0)
+Requires-Dist: litellm (==1.78.7)
+Requires-Dist: openai (>=1.106.0)
 Requires-Dist: opentelemetry-api (>=1.27.0,<2.0.0)
 Requires-Dist: opentelemetry-exporter-otlp-proto-http (>=1.27.0,<2.0.0)
 Requires-Dist: opentelemetry-instrumentation (>=0.56b0)
 Requires-Dist: opentelemetry-sdk (>=1.27.0,<2.0.0)
 Requires-Dist: pydantic (>=2,<3)
 Requires-Dist: python-dotenv (>=1.0.0,<2.0.0)
+Requires-Dist: python-jsonpath (>=2.0.0,<3.0.0)
 Requires-Dist: pyyaml (>=6.0.2,<7.0.0)
+Requires-Dist: restrictedpython (>=8.0,<9.0) ; python_version >= "3.11" and python_version < "3.14"
 Requires-Dist: starlette (>=0.47.0,<0.48.0)
 Requires-Dist: structlog (>=25.2.0,<26.0.0)
 Requires-Dist: tiktoken (==0.11.0)
 Requires-Dist: toml (>=0.10.2,<0.11.0)
-Project-URL: Documentation, https://docs.agenta.ai
+Project-URL: Documentation, https://agenta.ai/docs/
 Project-URL: Homepage, https://agenta.ai
 Project-URL: Repository, https://github.com/agenta-ai/agenta
 Description-Content-Type: text/markdown
@@ -45,11 +47,12 @@ Description-Content-Type: text/markdown
 <p align="center">
   <a href="https://agenta.ai?utm_source=github&utm_medium=referral&utm_campaign=readme">
       <picture >
-        <source width="275" media="(prefers-color-scheme: dark)" srcset="https://github.com/Agenta-AI/agenta/assets/4510758/cdddf5ad-2352-4920-b1d9-ae7f8d9d7735"  >
-        <source width="275" media="(prefers-color-scheme: light)" srcset="https://github.com/Agenta-AI/agenta/assets/4510758/ab75cbac-b807-496f-aab3-57463a33f726"  >
-        <img alt="Shows the logo of agenta" src="https://github.com/Agenta-AI/agenta/assets/4510758/68e055d4-d7b8-4943-992f-761558c64253" >
+        <source width="275" media="(prefers-color-scheme: dark)" srcset="https://github.com/user-attachments/assets/97e31bfc-b1fc-4d19-b443-5aedf6029017"  >
+        <source width="275" media="(prefers-color-scheme: light)" srcset="https://github.com/user-attachments/assets/fdc5f23f-2095-4cfc-9511-14c6851c1262"  >
+        <img alt="Shows the logo of agenta" src="https://github.com/user-attachments/assets/fdc5f23f-2095-4cfc-9511-14c6851c1262" >
       </picture>
   </a>
+
 <div align="center">
   <strong> <h1> The Open-source LLMOps Platform </h1></strong>
   Build reliable LLM applications faster with integrated prompt management, evaluation, and observability.
@@ -58,9 +61,9 @@ Description-Content-Type: text/markdown
       <div align="center" >
         <a href="https://cloud.agenta.ai?utm_source=github&utm_medium=referral&utm_campaign=readme">
           <picture >
-        <source media="(prefers-color-scheme: dark)" srcset="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/9a4fde42-c96c-4ec2-b8d1-c8c0e3851a00/large"  >
-        <source media="(prefers-color-scheme: light)" srcset="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/98140352-14c0-4db1-bafb-a1e8d271d500/large"  >
-        <img alt="Shows the logo of agenta" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/98140352-14c0-4db1-bafb-a1e8d271d500/large" >
+        <source media="(prefers-color-scheme: dark)" srcset="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/6fa19a9d-9785-4acf-5d08-e81b1e38b100/large"  >
+        <source media="(prefers-color-scheme: light)" srcset="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/6fa19a9d-9785-4acf-5d08-e81b1e38b100/large"  >
+        <img alt="Shows the logo of agenta" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/6fa19a9d-9785-4acf-5d08-e81b1e38b100/large" >
           </picture>
         </a>
     </div>
@@ -69,7 +72,7 @@ Description-Content-Type: text/markdown
 ---
 
 <h3 align="center">
-  <a href="https://docs.agenta.ai?utm_source=github&utm_medium=referral&utm_campaign=readme"><b>Documentation</b></a> &bull;
+  <a href="https://agenta.ai/docs/?utm_source=github&utm_medium=referral&utm_campaign=readme"><b>Documentation</b></a> &bull;
   <a href="https://agenta.ai?utm_source=github&utm_medium=referral&utm_campaign=readme"><b>Website</b></a> &bull;
   <a href="https://cloud.agenta.ai?utm_source=github&utm_medium=referral&utm_campaign=readme"><b>Agenta Cloud</b></a>
 </h3>
@@ -78,7 +81,7 @@ Description-Content-Type: text/markdown
 
 <p align="center">
   <img src="https://img.shields.io/badge/license-MIT-blue.svg" alt="MIT license." />
-  <a href="https://docs.agenta.ai?utm_source=github&utm_medium=referral&utm_campaign=readme">
+  <a href="https://agenta.ai/docs/?utm_source=github&utm_medium=referral&utm_campaign=readme">
     <img src="https://img.shields.io/badge/Doc-online-green" alt="Doc">
   </a>
   <a href="https://github.com/Agenta-AI/agenta/blob/main/CONTRIBUTING.md">
@@ -126,20 +129,20 @@ Agenta is a platform for building production-grade LLM applications. It helps **
 Collaborate with Subject Matter Experts (SMEs) on prompt engineering and make sure nothing breaks in production.
 
 - **Interactive Playground**: Compare prompts side by side against your test cases
-- **Multi-Model Support**: Experiment with 50+ LLM models or [bring-your-own models](https://docs.agenta.ai/prompt-engineering/playground/adding-custom-providers?utm_source=github&utm_medium=referral&utm_campaign=readme)
+- **Multi-Model Support**: Experiment with 50+ LLM models or [bring-your-own models](https://agenta.ai/docs/prompt-engineering/playground/custom-providers?utm_source=github&utm_medium=referral&utm_campaign=readme)
 - **Version Control**: Version prompts and configurations with branching and environments
-- **Complex Configurations**: Enable SMEs to collaborate on [complex configuration schemas](https://docs.agenta.ai/custom-workflows/overview?utm_source=github&utm_medium=referral&utm_campaign=readme) beyond simple prompts
+- **Complex Configurations**: Enable SMEs to collaborate on [complex configuration schemas](https://agenta.ai/docs/custom-workflows/overview?utm_source=github&utm_medium=referral&utm_campaign=readme) beyond simple prompts
 
-[Explore prompt management →](https://docs.agenta.ai/prompt-engineering/overview?utm_source=github&utm_medium=referral&utm_campaign=readme)
+[Explore prompt management →](https://agenta.ai/docs/prompt-engineering/concepts?utm_source=github&utm_medium=referral&utm_campaign=readme)
 
 ### 📊 Evaluation & Testing
 Evaluate your LLM applications systematically with both human and automated feedback.
-- **Flexible Test Sets**: Create test cases from production data, playground experiments, or upload CSVs
+- **Flexible Testsets**: Create testcases from production data, playground experiments, or upload CSVs
 - **Pre-built and Custom Evaluators**: Use LLM-as-judge, one of our 20+ pre-built evaluators, or you custom evaluators
 - **UI and API Access**: Run evaluations via UI (for SMEs) or programmatically (for engineers)
 - **Human Feedback Integration**: Collect and incorporate expert annotations
 
-[Explore evaluation frameworks →](https://docs.agenta.ai/evaluation/overview?utm_source=github&utm_medium=referral&utm_campaign=readme)
+[Explore evaluation frameworks →](https://agenta.ai/docs/evaluation/overview?utm_source=github&utm_medium=referral&utm_campaign=readme)
 
 ### 📡 Observability & Monitoring
 Get visibility into your LLM applications in production.
@@ -148,14 +151,14 @@ Get visibility into your LLM applications in production.
 - **Open Standards**: OpenTelemetry native tracing compatible with OpenLLMetry, and OpenInference
 - **Integrations**: Comes with pre-built integrations for most models and frameworks
 
-[Learn about observability →](https://docs.agenta.ai/observability/overview?utm_source=github&utm_medium=referral&utm_campaign=readme)
+[Learn about observability →](https://agenta.ai/docs/observability/overview?utm_source=github&utm_medium=referral&utm_campaign=readme)
 
 ## 📸 Screenshots
 
-<img alt="Playground" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/e7d16967-d564-4148-3615-03bfe2a29d00/large" />
-<img alt="Prompt Management" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/b4210a13-b41e-4dfd-d383-4286fb011e00/large" />
-<img alt="Evaluation" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/4b1c62b9-ff6e-4bd1-4c8d-de59b8485e00/large" />
-<img  alt="Observability" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/5c764759-e525-499e-544a-1077c45e9600/large" />
+<img alt="Playground" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/a4f67ac4-1acc-40c6-7a1a-5616eee7bb00/large" />
+<img alt="Prompt Management" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/65f697d0-3221-4e3c-7232-f350b1976a00/large" />
+<img alt="Evaluation" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/19b5b77e-6945-4419-15b9-cfea197e1300/large" />
+<img  alt="Observability" src="https://imagedelivery.net/UNvjPBCIZFONpkVPQTxVuA/efc8a24c-2a2a-427c-f285-7d8b41200700/large" />
 
 ## 🚀 Getting Started 
 
@@ -188,7 +191,7 @@ docker compose -f hosting/docker-compose/oss/docker-compose.gh.yml --env-file ho
 
 3. Access Agenta at `http://localhost`.
 
-For deploying on a remote host, or using different ports refers to our [self-hosting](https://docs.agenta.ai/self-host/quick-start?utm_source=github&utm_medium=referral&utm_campaign=readme) and [remote deployment documentation](https://docs.agenta.ai/self-host/guides/deploy-remotely?utm_source=github&utm_medium=referral&utm_campaign=readme).
+For deploying on a remote host, or using different ports refers to our [self-hosting](https://agenta.ai/docs/self-host/quick-start?utm_source=github&utm_medium=referral&utm_campaign=readme) and [remote deployment documentation](https://agenta.ai/docs/self-host/guides/deploy-remotely?utm_source=github&utm_medium=referral&utm_campaign=readme).
 
 ## 💬 Community
 
@@ -196,8 +199,8 @@ Find help, explore resources, or get involved:
 
 ### 🧰 Support
 
-- **📚 [Documentation](https://docs.agenta.ai?utm_source=github&utm_medium=referral&utm_campaign=readme)** – Full guides and API reference  
-- **📋 [Changelog](https://docs.agenta.ai/changelog/main?utm_source=github&utm_medium=referral&utm_campaign=readme)** – Track recent updates  
+- **📚 [Documentation](https://agenta.ai/docs/?utm_source=github&utm_medium=referral&utm_campaign=readme)** – Full guides and API reference  
+- **📋 [Changelog](https://agenta.ai/docs/changelog/main?utm_source=github&utm_medium=referral&utm_campaign=readme)** – Track recent updates  
 - **💬 [Slack Community](https://join.slack.com/t/agenta-hq/shared_invite/zt-37pnbp5s6-mbBrPL863d_oLB61GSNFjw)** – Ask questions and get support  
 
 ### 🤝 Contribute
@@ -206,7 +209,7 @@ We welcome contributions of all kinds — from filing issues and sharing ideas t
 
 - **🐛 [Report bugs](https://github.com/Agenta-AI/agenta/issues)** – Help us by reporting problems you encounter  
 - **💡 [Share ideas and feedback](https://github.com/Agenta-AI/agenta/discussions)** – Suggest features or vote on ideas
-- **🔧 [Contribute to the codebase](https://docs.agenta.ai/misc/contributing/getting-started?utm_source=github&utm_medium=referral&utm_campaign=readme)** – Read the guide and open a pull request
+- **🔧 [Contribute to the codebase](https://agenta.ai/docs/misc/contributing/getting-started?utm_source=github&utm_medium=referral&utm_campaign=readme)** – Read the guide and open a pull request
 
 ## ⭐ Star Agenta
 
@@ -224,7 +227,7 @@ We welcome contributions of all kinds — from filing issues and sharing ideas t
 ## Contributors ✨
 
 <!-- ALL-CONTRIBUTORS-BADGE:START - Do not remove or modify this section -->
-[![All Contributors](https://img.shields.io/badge/all_contributors-49-orange.svg?style=flat-square)](#contributors-)
+[![All Contributors](https://img.shields.io/badge/all_contributors-50-orange.svg?style=flat-square)](#contributors-)
 <!-- ALL-CONTRIBUTORS-BADGE:END -->
 
 Thanks goes to these wonderful people ([emoji key](https://allcontributors.org/docs/en/emoji-key)):