mockstack 0.7.3__tar.gz → 0.9.0__tar.gz

{mockstack-0.7.3 → mockstack-0.9.0}/Dockerfile

@@ -2,7 +2,7 @@
 
 # Define build argument for version
 ARG PYTHON_IMAGE_VERSION=3.13.3-slim
-ARG MOCKSTACK_VERSION=0.1.0
+ARG MOCKSTACK_VERSION=0.8.0
 
 # ---------------------------- Base Image --------------------------------
 

{mockstack-0.7.3 → mockstack-0.9.0}/PKG-INFO

@@ -1,11 +1,11 @@
 Metadata-Version: 2.4
 Name: mockstack
-Version: 0.7.3
+Version: 0.9.0
 Summary: An API mocking workhorse
 Author-email: Adam Ever-Hadani <mockstack.contact@gmail.com>
 License-Expression: MIT
-Project-URL: Homepage, https://github.com/adamhadani/mockstack
-Project-URL: Issues, https://github.com/adamhadani/mockstack/issues
+Project-URL: Homepage, https://github.com/promptromp/mockstack
+Project-URL: Issues, https://github.com/promptromp/mockstack/issues
 Keywords: mocking,integration-testing,testing,microservices,api
 Classifier: Development Status :: 5 - Production/Stable
 Classifier: Programming Language :: Python
@@ -27,15 +27,17 @@ Requires-Dist: pyyaml>=6.0.2
 Requires-Dist: rich>=14.0.0
 Requires-Dist: typer>=0.15.2
 Requires-Dist: uvicorn>=0.34.2
+Provides-Extra: llm
+Requires-Dist: ollama>=0.4.8; extra == "llm"
 Dynamic: license-file
 
-![mockstack logo](https://github.com/adamhadani/mockstack/raw/main/docs/assets/mockstack.png)
+![mockstack logo](https://github.com/promptromp/mockstack/raw/main/docs/assets/mockstack.png)
 
 --------------------------------------------------------------------------------
 
 
-[![CI](https://github.com/adamhadani/mockstack/actions/workflows/ci.yml/badge.svg)](https://github.com/adamhadani/mockstack/actions/workflows/ci.yml)
-[![GitHub License](https://img.shields.io/github/license/adamhadani/mockstack)](https://github.com/adamhadani/mockstack/blob/main/LICENSE)
+[![CI](https://github.com/promptromp/mockstack/actions/workflows/ci.yml/badge.svg)](https://github.com/promptromp/mockstack/actions/workflows/ci.yml)
+[![GitHub License](https://img.shields.io/github/license/promptromp/mockstack)](https://github.com/promptromp/mockstack/blob/main/LICENSE)
 [![PyPI - Version](https://img.shields.io/pypi/v/mockstack)](https://pypi.org/project/mockstack/)
 [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/mockstack)](https://pypi.org/project/mockstack/)
 
@@ -50,7 +52,7 @@ Use mockstack for:
 
 * **Integration Testing** :ok_hand:. Creating a consistent environment for running integration tests on a single component, using fixture responses.
 
-* **LLM-powered Workflows** :alien:. Speeding up development of LLM-based workflows and tools for use with frameworks such as [LangChain](https://python.langchain.com/docs/introduction/), [LangGraph](https://www.langchain.com/langgraph) and others. When developing LLM-driven execution graphs, you often want to have a consistent response from a tool while you're tuning prompts or debugging other aspects of a particular trace. **mockstack** can give you a solid foundation for this.
+* **LLM-powered Workflows** :alien:. Speeding up development and reducing per-token costs of LLM-based workflows and tools for use with frameworks such as [LangChain](https://python.langchain.com/docs/introduction/), [LangGraph](https://www.langchain.com/langgraph) and others. When developing LLM-driven execution graphs. Optional [Ollama](https://ollama.com/) integration allows for realistic mocking of 3rd party LLMs without changing a line of code in your project. In addition, when you want to have a consistent response from a tool while you're tuning prompts or debugging other aspects of a particular trace you can create various fixture responses with varying levels of dynamic content that is template-driven. **mockstack** can give you a solid foundation for this.
 
 * **Chaos Engineering** :boom:. mockstack can simulate various real-world runtime error scenarios such as timeouts, http error codes, and invalid response payloads. This can be a great way to do some upfront [Chaos Monkey](https://github.com/Netflix/chaosmonkey) type of testing on software components.
 
@@ -75,21 +77,28 @@ or install into a persistent environment and add it to the PATH with:
 
 ## Usage
 
-See the [examples](https://github.com/adamhadani/mockstack/blob/main/examples/) directory for complete examples with documentation.
+See the [examples](https://github.com/promptromp/mockstack/blob/main/examples/) directory for complete examples with documentation.
 
-Available configuration options are [here](https://github.com/adamhadani/mockstack/blob/main/mockstack/config.py).
+Available configuration options are [here](https://github.com/promptromp/mockstack/blob/main/mockstack/config.py).
 
-Setting individual options can be done either through an `.env` file, individual environment variables, or command-line arguments. For example:
+Setting individual options can be done either through an `.env` file, individual environment variables, or command-line arguments.
+
+
+Minimal example to get you started:
 
 ```shell
-    export MOCKSTACK__STRATEGY=filefixtures
+    mkdir -p ~/mockstack-templates
+    echo '{"message": "Hello from mockstack!"}' > ~/mockstack-templates/myservice-api-myresource.j2
+
     export MOCKSTACK__TEMPLATES_DIR=~/mockstack-templates/
-    export MOCKSTACK__OPENTELEMETRY__ENABLED=true
-    export MOCKSTACK__OPENTELEMETRY__CAPTURE_RESPONSE_BODY=true
     uvx mockstack
 ```
 
-See also the included [.env.example](https://github.com/adamhadani/mockstack/blob/main/.env.example) for more examples. You can copy that file to `.env` and fill in configuration as needed based on the given examples.
+You can then hit `http://localhost:8000/myservice/api/myresource/23faa8cf-5daa-4bcb-8c92-27018b712aa9` (or any other UUID).
+
+This is of course just the tip of the iceberg.
+
+See also the included [.env.example](https://github.com/promptromp/mockstack/blob/main/.env.example) for more settings you are likely to find useful. You can copy that file to `.env` and fill in configuration as needed based on the given examples.
 
 Out of the box, you get the following behavior when using the default `filefixtures` strategy:
 
@@ -118,7 +127,7 @@ Linting, formatting, static type checks etc. are all managed via [pre-commit](ht
 
 If you are contributing to development, you will want to clone this project, and can then install it locally with:
 
-    gh repo clone adamhadani/mockstack
+    gh repo clone promptromp/mockstack
     cd mockstack/
     uv sync
     uv pip install -e .

{mockstack-0.7.3 → mockstack-0.9.0}/README.md

@@ -1,10 +1,10 @@
-![mockstack logo](https://github.com/adamhadani/mockstack/raw/main/docs/assets/mockstack.png)
+![mockstack logo](https://github.com/promptromp/mockstack/raw/main/docs/assets/mockstack.png)
 
 --------------------------------------------------------------------------------
 
 
-[![CI](https://github.com/adamhadani/mockstack/actions/workflows/ci.yml/badge.svg)](https://github.com/adamhadani/mockstack/actions/workflows/ci.yml)
-[![GitHub License](https://img.shields.io/github/license/adamhadani/mockstack)](https://github.com/adamhadani/mockstack/blob/main/LICENSE)
+[![CI](https://github.com/promptromp/mockstack/actions/workflows/ci.yml/badge.svg)](https://github.com/promptromp/mockstack/actions/workflows/ci.yml)
+[![GitHub License](https://img.shields.io/github/license/promptromp/mockstack)](https://github.com/promptromp/mockstack/blob/main/LICENSE)
 [![PyPI - Version](https://img.shields.io/pypi/v/mockstack)](https://pypi.org/project/mockstack/)
 [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/mockstack)](https://pypi.org/project/mockstack/)
 
@@ -19,7 +19,7 @@ Use mockstack for:
 
 * **Integration Testing** :ok_hand:. Creating a consistent environment for running integration tests on a single component, using fixture responses.
 
-* **LLM-powered Workflows** :alien:. Speeding up development of LLM-based workflows and tools for use with frameworks such as [LangChain](https://python.langchain.com/docs/introduction/), [LangGraph](https://www.langchain.com/langgraph) and others. When developing LLM-driven execution graphs, you often want to have a consistent response from a tool while you're tuning prompts or debugging other aspects of a particular trace. **mockstack** can give you a solid foundation for this.
+* **LLM-powered Workflows** :alien:. Speeding up development and reducing per-token costs of LLM-based workflows and tools for use with frameworks such as [LangChain](https://python.langchain.com/docs/introduction/), [LangGraph](https://www.langchain.com/langgraph) and others. When developing LLM-driven execution graphs. Optional [Ollama](https://ollama.com/) integration allows for realistic mocking of 3rd party LLMs without changing a line of code in your project. In addition, when you want to have a consistent response from a tool while you're tuning prompts or debugging other aspects of a particular trace you can create various fixture responses with varying levels of dynamic content that is template-driven. **mockstack** can give you a solid foundation for this.
 
 * **Chaos Engineering** :boom:. mockstack can simulate various real-world runtime error scenarios such as timeouts, http error codes, and invalid response payloads. This can be a great way to do some upfront [Chaos Monkey](https://github.com/Netflix/chaosmonkey) type of testing on software components.
 
@@ -44,21 +44,28 @@ or install into a persistent environment and add it to the PATH with:
 
 ## Usage
 
-See the [examples](https://github.com/adamhadani/mockstack/blob/main/examples/) directory for complete examples with documentation.
+See the [examples](https://github.com/promptromp/mockstack/blob/main/examples/) directory for complete examples with documentation.
 
-Available configuration options are [here](https://github.com/adamhadani/mockstack/blob/main/mockstack/config.py).
+Available configuration options are [here](https://github.com/promptromp/mockstack/blob/main/mockstack/config.py).
 
-Setting individual options can be done either through an `.env` file, individual environment variables, or command-line arguments. For example:
+Setting individual options can be done either through an `.env` file, individual environment variables, or command-line arguments.
+
+
+Minimal example to get you started:
 
 ```shell
-    export MOCKSTACK__STRATEGY=filefixtures
+    mkdir -p ~/mockstack-templates
+    echo '{"message": "Hello from mockstack!"}' > ~/mockstack-templates/myservice-api-myresource.j2
+
     export MOCKSTACK__TEMPLATES_DIR=~/mockstack-templates/
-    export MOCKSTACK__OPENTELEMETRY__ENABLED=true
-    export MOCKSTACK__OPENTELEMETRY__CAPTURE_RESPONSE_BODY=true
     uvx mockstack
 ```
 
-See also the included [.env.example](https://github.com/adamhadani/mockstack/blob/main/.env.example) for more examples. You can copy that file to `.env` and fill in configuration as needed based on the given examples.
+You can then hit `http://localhost:8000/myservice/api/myresource/23faa8cf-5daa-4bcb-8c92-27018b712aa9` (or any other UUID).
+
+This is of course just the tip of the iceberg.
+
+See also the included [.env.example](https://github.com/promptromp/mockstack/blob/main/.env.example) for more settings you are likely to find useful. You can copy that file to `.env` and fill in configuration as needed based on the given examples.
 
 Out of the box, you get the following behavior when using the default `filefixtures` strategy:
 
@@ -87,7 +94,7 @@ Linting, formatting, static type checks etc. are all managed via [pre-commit](ht
 
 If you are contributing to development, you will want to clone this project, and can then install it locally with:
 
-    gh repo clone adamhadani/mockstack
+    gh repo clone promptromp/mockstack
     cd mockstack/
     uv sync
     uv pip install -e .

{mockstack-0.7.3 → mockstack-0.9.0}/SECURITY.md

@@ -11,4 +11,4 @@ We do not currently run any bug bounty programs.
 ## Vulnerability disclosures
 
 Critical vulnerabilities will be disclosed via GitHub's
-[security advisory](https://github.com/adamhadani/mockstack/security) system.
+[security advisory](https://github.com/promptromp/mockstack/security) system.

mockstack-0.9.0/docs/ollama.md

@@ -0,0 +1,87 @@
+# Ollama Integration
+
+Mockstack provides integration with [Ollama](https://ollama.com/), allowing you to use real LLM responses in your mock templates. This is particularly useful for development, debugging, and integration testing scenarios where you want to capture the non-deterministic nature of LLM responses.
+
+## Prerequisites
+
+To use the Ollama integration, you'll need:
+
+1. Mockstack installed with the optional `llm` dependencies:
+   ```bash
+   uv pip install mockstack[llm]
+   ```
+
+2. [Ollama](https://ollama.com/) installed locally with at least one model (e.g., "llama3.2")
+
+## Basic Usage
+
+The Ollama integration works by routing requests to a template file that uses the special `ollama` template function.
+
+1. Configure your LLM client to hit an endpoint that maps to a template filepath calling the `ollama` method, e.g.:
+   ```python
+   from langchain_openai import ChatOpenAI
+
+   llm = ChatOpenAI(
+       model="gpt-4o",
+       base_url="http://localhost:8000/ollama/openai/v1",
+       api_key="SOME_STRING_THAT_DOES_NOT_MATTER",
+   )
+   ```
+
+2. Make requests as you normally would:
+   ```python
+   messages = [
+       (
+           "system",
+           "You are a helpful assistant that translates English to French. Translate the user sentence.",
+       ),
+       ("human", "mockstack is pretty cool. But LLMs are way cooler"),
+   ]
+   ai_msg = llm.invoke(messages)
+   print(ai_msg.content)
+   ```
+
+## Integration with Templates
+
+You can use Ollama responses within your Jinja templates via the provided `ollama` template function. This allows you to:
+
+1. Mix static and dynamic content
+2. Apply transformations to the LLM responses
+3. Create conditional logic based on the responses
+
+Example template structure:
+
+```jinja
+{
+  "id": "chatcmpl-123",
+  "object": "chat.completion",
+  "created": 1677652288,
+  "model": "gpt-4.1",
+  "choices": [{
+    "index": 0,
+    "message": {
+      "role": "assistant",
+      "content": "{{ ollama(request_json.messages, 'llama3.2') | json_escape }}"
+    },
+    "finish_reason": "stop"
+  }],
+  "usage": {
+    "prompt_tokens": 9,
+    "completion_tokens": 12,
+    "total_tokens": 21
+  }
+}
+```
+
+## Best Practices
+
+1. **Caching**: Consider implementing caching for frequently used responses to improve performance
+2. **Model Selection**: Choose the appropriate model based on your testing needs
+3. **Error Handling**: Implement proper error handling in your templates
+4. **Performance**: Be mindful of response times when using real LLM responses
+
+## Limitations
+
+- The integration requires a local Ollama instance
+- Response times will be slower than static templates
+- Model availability depends on your local setup

{mockstack-0.7.3 → mockstack-0.9.0}/examples/llm/mockstack-langchain-example.ipynb

@@ -11,7 +11,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -22,11 +22,31 @@
     "#!uv pip install langchain langchain-openai"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from ollama import chat\n",
+    "from ollama import ChatResponse\n",
+    "\n",
+    "response: ChatResponse = chat(model='llama3.2', messages=[\n",
+    "  {\n",
+    "    'role': 'user',\n",
+    "    'content': 'Why is the sky blue?',\n",
+    "  },\n",
+    "])\n",
+    "print(response['message']['content'])\n",
+    "# or access fields directly from the response object\n",
+    "print(response.message.content)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Example #1: Template-based mocking (`filefixtures` strategy)\n",
+    "## Example #1: Static template-based mocking (`filefixtures` strategy)\n",
     "\n",
     "Here we simply use the **filefixtures** strategy to route requesets coming in for a certain URL to a template file with the appropriate name.\n",
     "\n",
@@ -42,7 +62,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 38,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -86,7 +106,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 40,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -120,17 +140,62 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Example #2: `llm` strategy\n",
+    "## Example #2: Dynamic template-based mocking with ollama integration\n",
     "\n",
-    "** COMING SOON **"
+    "In this example we use the [ollama-python](https://github.com/ollama/ollama-python) integration to return mock responses that actually come from a real LLM running on your host! This is a great way to go a step further in development, debugging, and integration testing scenarios where LLMs and their non-determinism are important to capture.\n",
+    "\n",
+    "For this example you will need to make sure you have the following:\n",
+    "\n",
+    "* mockstack installed with the optional `llm` dependencies:\n",
+    "\n",
+    "    ```bash\n",
+    "    uv pip install mockstack[llm]\n",
+    "    ```\n",
+    "\n",
+    "* [ollama](https://ollama.com/) installed locally with the \"llama3.2\" model (which is typically the default model installed)"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": null,
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Le mockstack est très amusant, mais les LLMs sont beaucoup plus froids.\n",
+      "\n",
+      "(Note: I used \"froids\" instead of \"cool\" as it's a more common translation in French. If you want to keep the exact word order and meaning, I can suggest an alternative: Le mockstack est très amusant, mais les LLMs sont vraiment cool.)\n"
+     ]
+    }
+   ],
+   "source": [
+    "from langchain_openai import ChatOpenAI\n",
+    "\n",
+    "\n",
+    "llm = ChatOpenAI(\n",
+    "    model=\"gpt-4o\",\n",
+    "    temperature=0,\n",
+    "    max_tokens=None,\n",
+    "    timeout=None,\n",
+    "    max_retries=2,\n",
+    "    base_url=\"http://localhost:8000/ollama/openai/v1\",\n",
+    "    api_key=\"SOME_STRING_THAT_DOES_NOT_MATTER\",\n",
+    ")\n",
+    "\n",
+    "messages = [\n",
+    "    (\n",
+    "        \"system\",\n",
+    "        \"You are a helpful assistant that translates English to French. Translate the user sentence.\",\n",
+    "    ),\n",
+    "    (\"human\", \"mockstack is pretty cool. But LLMs are way cooler\"),\n",
+    "]\n",
+    "ai_msg = llm.invoke(messages)\n",
+    "\n",
+    "# Output would not be deterministic, but should be a valid French translation.\n",
+    "print(ai_msg.content)"
+   ]
   },
   {
    "cell_type": "markdown",

mockstack-0.9.0/examples/llm/templates/ollama-openai-v1-chat-completions.j2

@@ -0,0 +1,35 @@
+ {
+   "id": "chatcmpl-B9MBs8CjcvOU2jLn4n570S5qMJKcT",
+   "object": "chat.completion",
+   "created": 1741569952,
+   "model": "gpt-4.1-2025-04-14",
+   "choices": [
+     {
+       "index": 0,
+       "message": {
+         "role": "assistant",
+         "content": "{{ ollama(request_json.messages, 'llama3.2') | json_escape }}",
+         "refusal": null,
+         "annotations": []
+       },
+       "logprobs": null,
+       "finish_reason": "stop"
+     }
+   ],
+   "usage": {
+     "prompt_tokens": 19,
+     "completion_tokens": 10,
+     "total_tokens": 29,
+     "prompt_tokens_details": {
+       "cached_tokens": 0,
+       "audio_tokens": 0
+     },
+     "completion_tokens_details": {
+       "reasoning_tokens": 0,
+       "audio_tokens": 0,
+       "accepted_prediction_tokens": 0,
+       "rejected_prediction_tokens": 0
+     }
+   },
+   "service_tier": "default"
+ }

{mockstack-0.7.3 → mockstack-0.9.0}/mkdocs.yml

@@ -1,7 +1,7 @@
 site_name: mockstack
-site_url: https://adamhadani.github.io/mockstack
+site_url: https://promptromp.github.io/mockstack
 repo_name: mockstack
-repo_url: https://github.com/adamhadani/mockstack
+repo_url: https://github.com/promptromp/mockstack
 
 theme:
   name: material
@@ -49,10 +49,19 @@ markdown_extensions:
   - pymdownx.tabbed:
       alternate_style: true
 
+nav:
+  - Home: README.md
+  - Configuration: configuration.md
+  - Strategies:
+    - BaseStrategy: strategies/base.md
+    - FileFixtures: strategies/filefixtures.md
+    - ProxyRules: strategies/proxyrules.md
+  - LLM Integrations:
+    - Ollama: ollama.md
 
 extra:
   social:
     - icon: simple/github
-      link: https://github.com/adamhadani/mockstack
+      link: https://github.com/promptromp/mockstack
 
 copyright: Copyright © 2025 Adam Ever-Hadani

{mockstack-0.7.3 → mockstack-0.9.0}/mockstack/config.py

@@ -20,13 +20,13 @@ from mockstack.constants import (
 class OpenTelemetrySettings(BaseSettings):
     """Settings for OpenTelemetry."""
 
-    enabled: bool = False
+    enabled: CliImplicitFlag[bool] = False
 
     endpoint: str = "http://localhost:4317/"
 
     # whether to capture the response body.
     # this can be heavy, sensitive (PII) and/or not needed depending on the use case.
-    capture_response_body: bool = False
+    capture_response_body: CliImplicitFlag[bool] = False
 
 
 class Settings(BaseSettings):

mockstack-0.9.0/mockstack/exceptions.py

@@ -0,0 +1,6 @@
+"""Custom exceptions for mockstack."""
+
+
+def raise_for_missing(message: str, *args, **kwargs):
+    """Raise an exception for a missing dependency."""
+    raise RuntimeError(message)

mockstack-0.9.0/mockstack/llm/ollama.py

@@ -0,0 +1,50 @@
+"""Ollama integration"""
+
+from typing import List, Dict
+
+try:
+    from ollama import chat
+    from ollama import ChatResponse
+
+    IS_OLLAMA_AVAILABLE = True
+except ImportError:
+    IS_OLLAMA_AVAILABLE = False
+
+
+if IS_OLLAMA_AVAILABLE:
+
+    class OllamaLLM:
+        def __init__(self, model: str = "llama3.2"):
+            self.model = model
+
+        def __call__(
+            self,
+            messages: List[Dict[str, str]],
+            max_tokens: int = 4096,
+            temperature: float = 0.7,
+        ) -> str:
+            response: ChatResponse = chat(
+                model=self.model,
+                messages=messages,
+                options={"num_ctx": max_tokens, "temperature": temperature},
+            )
+
+            return content(response)
+
+    def content(response: ChatResponse) -> str:
+        """Extract the message content from the LLM response."""
+        return response["message"]["content"]
+
+    def ollama(
+        messages: List[Dict[str, str]],
+        model: str = "llama3.2",
+        *args,
+        **kwargs,
+    ) -> str:
+        """Fluent interface for Ollama to be used in templates."""
+
+        return OllamaLLM(model)(
+            messages,
+            *args,
+            **kwargs,
+        )

{mockstack-0.7.3 → mockstack-0.9.0}/mockstack/strategies/filefixtures.py

@@ -2,21 +2,24 @@
 
 import logging
 import os
+from functools import cached_property
 from pathlib import Path
 
 from fastapi import HTTPException, Request, Response, status
-from jinja2 import Environment, FileSystemLoader
+from fastapi.responses import JSONResponse
+from jinja2 import Environment
 
 from mockstack.config import Settings
 from mockstack.intent import (
     looks_like_a_command,
     looks_like_a_search,
+    wants_json,
 )
 from mockstack.strategies.base import BaseStrategy
 from mockstack.strategies.create_mixin import CreateMixin
 from mockstack.templating import (
     iter_possible_template_arguments,
-    missing_template_detail,
+    templates_env_provider,
 )
 
 
@@ -37,8 +40,6 @@ class FileFixturesStrategy(BaseStrategy, CreateMixin):
         self.created_resource_metadata = settings.created_resource_metadata
         self.missing_resource_fields = settings.missing_resource_fields
 
-        self.env = Environment(loader=FileSystemLoader(self.templates_dir))
-
     def __str__(self) -> str:
         return (
             f"[medium_purple]filefixtures[/medium_purple]\n "
@@ -46,6 +47,11 @@ class FileFixturesStrategy(BaseStrategy, CreateMixin):
             f"enable_templates_for_post: [medium_purple]{self.enable_templates_for_post}[/medium_purple]. "
         )
 
+    @cached_property
+    def env(self) -> Environment:
+        """Jinja2 environment for the filefixtures strategy."""
+        return templates_env_provider(self.templates_dir)
+
     async def apply(self, request: Request) -> Response:
         match request.method:
             case "GET":
@@ -74,9 +80,10 @@ class FileFixturesStrategy(BaseStrategy, CreateMixin):
         We also allow a configuration to specify a default intent.
 
         """
+        request_json = (await request.json()) if wants_json(request) else None
         if self.enable_templates_for_post:
             try:
-                return self._response_from_template(request)
+                return self._response_from_template(request, request_json=request_json)
             except HTTPException as e:
                 if e.status_code == status.HTTP_404_NOT_FOUND:
                     # If the template is not found, we try to create the resource with logic below.
@@ -86,12 +93,12 @@ class FileFixturesStrategy(BaseStrategy, CreateMixin):
 
         if looks_like_a_search(request):
             # Searching for resources with a complex query that cannot be expressed in a URI.
-            return self._response_from_template(request)
+            return self._response_from_template(request, request_json=request_json)
         elif looks_like_a_command(request):
             # Executing a 'command' of some sort, like a workflow or a batch job.
             # We return a 201 CREATED status code with response from template.
             return self._response_from_template(
-                request, status_code=status.HTTP_201_CREATED
+                request, request_json=request_json, status_code=status.HTTP_201_CREATED
             )
         else:
             # simulate resource creation:
@@ -133,9 +140,15 @@ class FileFixturesStrategy(BaseStrategy, CreateMixin):
         return Response(status_code=status.HTTP_204_NO_CONTENT)
 
     def _response_from_template(
-        self, request: Request, status_code: int = status.HTTP_200_OK
+        self,
+        request: Request,
+        *,
+        request_json: dict | None = None,
+        status_code: int = status.HTTP_200_OK,
     ) -> Response:
-        for template_args in iter_possible_template_arguments(request):
+        for template_args in iter_possible_template_arguments(
+            request, request_json=request_json
+        ):
             filename = self.templates_dir / template_args["name"]
             self.logger.debug("Looking for template filename: %s", filename)
             if not os.path.exists(filename):
@@ -152,17 +165,10 @@ class FileFixturesStrategy(BaseStrategy, CreateMixin):
             )
 
         # if we get here, we have no template to render.
-        raise HTTPException(
-            status_code=status.HTTP_404_NOT_FOUND,
-            detail=missing_template_detail(request, templates_dir=self.templates_dir),
-        )
-        """
-        # TODO: return custom fields from settings
         return JSONResponse(
             content=self.missing_resource_fields,
             status_code=status.HTTP_404_NOT_FOUND,
         )
-        """
 
     def update_opentelemetry(self, request: Request, template_args: dict) -> None:
         """Update the opentelemetry span with the file fixtures details."""