update testing links

Author: lbeurerkellner

docs/testing/examples/langgraph.md

@@ -19,7 +19,7 @@ pip install langgraph
 
 ## Agent code
 
-You can view the agent code [here](https://github.com/invariantlabs-ai/invariant/blob/main/testing/sample_tests/langgraph/weather_agent/weather_agent.py).
+You can view the agent code [here](https://github.com/invariantlabs-ai/testing/blob/main/invariant_testing/testing/sample_tests/langgraph/weather_agent/weather_agent.py).
 
 This can be invoked as:
 

docs/testing/examples/swarm.md

@@ -19,11 +19,11 @@ To get started quickly with testing your Swarm agent, you can use the following
 from swarm import Agent, Swarm
 
 # for assertions
-from invariant.testing import assert_true
-from invariant.testing import functional as F
+from invariant_testing.testing import assert_true
+from invariant_testing.testing import functional as F
 
 # for swarm testing support
-from invariant.testing.wrappers.swarm_wrapper import SwarmWrapper
+from invariant_testing.testing.wrappers.swarm_wrapper import SwarmWrapper
 
 
 def get_capital(country_name: str) -> str:
@@ -88,7 +88,7 @@ pip install openai-swarm
 
 ## Agent code
 
-You can view the full code example of the example agent [here](https://github.com/invariantlabs-ai/invariant/blob/main/invariant/testing/sample_tests/swarm/capital_finder_agent/capital_finder_agent.py)
+You can view the full code example of the example agent [here](https://github.com/invariantlabs-ai/testing/blob/main/invariant_testing/testing/sample_tests/swarm/capital_finder_agent/capital_finder_agent.py)
 
 The agent can be invoked as follows.
 

docs/testing/index.md

@@ -4,7 +4,7 @@ title: Overview
 
 # Invariant `testing`: helps you build better AI agents through debuggable unit testing
 
-Invariant `testing` is a lightweight library to write and run AI agent tests. It provides helpers and assertions that enable you to write robust tests for your agentic applications.
+Invariant [`testing`](https://github.com/invariantlabs-ai/testing) is a lightweight library to write and run AI agent tests. It provides helpers and assertions that enable you to write robust tests for your agentic applications.
 
 Using [**localized assertions**](writing/traces.ipynb), `testing` always points you to the exact part of the agent's behavior that caused a test to fail, making it easy to debug and resolve issues (_think: stacktraces for agents_).
 
@@ -15,7 +15,7 @@ Using [**localized assertions**](writing/traces.ipynb), `testing` always points
 ## Installation
 
 ```
-pip install invariant-ai
+pip install git+https://github.com/invariantlabs-ai/testing.git
 ```
 
 ## A quick example
@@ -30,8 +30,8 @@ export OPENAI_API_KEY=<your-key>
 
 ```python
 # content of tests/test_weather.py
-import invariant.testing.functional as F
-from invariant.testing import Trace, assert_equals
+import invariant_testing.testing.functional as F
+from invariant_testing.testing import Trace, assert_equals
 
 def test_weather():
     # create a Trace object from your agent trajectory

docs/testing/quickstart.md

@@ -23,7 +23,7 @@ To get started, first create a `tests/` folder in your project root. This folder
 Next, create a new file in `tests/` and name it `test_example.py` as shown below:
 
 ```python
-from invariant.testing import Trace, assert_true
+from invariant_testing.testing import Trace, assert_true
 
 def test_weather():
     # create a Trace object from your agent trajectory

docs/testing/writing/matchers.md

@@ -13,18 +13,18 @@ To accommodate this, `testing` includes several different `Matcher` implementati
 
 Beyond that, `Matcher` is also a simple base class that allows you to write your own custom matchers, if the provided ones are not sufficient for your needs (e.g. custom properties).
 
-## [`IsSimilar`](https://github.com/invariantlabs-ai/invariant/blob/main/testing/invariant/custom_types/matchers.py#L53)
+## [`IsSimilar`](https://github.com/invariantlabs-ai/testing/blob/main/invariant_testing/testing/custom_types/matchers.py#L53)
 
 Checks if a string is similar to an expected string by checking if the similary score reaches a given threshold.
 
-## [`LambdaMatcher`](https://github.com/invariantlabs-ai/invariant/blob/main/testing/invariant/custom_types/matchers.py#L18)
+## [`LambdaMatcher`](https://github.com/invariantlabs-ai/testing/blob/main/invariant_testing/testing/custom_types/matchers.py#L18)
 
 Matcher for checking if a lambda function returns True on the underlying value. This can be useful to check for custom properties of outputs, while maintaining [addresses to localize failing](./tests.md) assertions.
 
-## [`IsFactuallyEqual`](https://github.com/invariantlabs-ai/invariant/blob/main/testing/invariant/custom_types/matchers.py#L86)
+## [`IsFactuallyEqual`](https://github.com/invariantlabs-ai/testing/blob/main/invariant_testing/testing/custom_types/matchers.py#L86)
 
 Checks for factual equality / entailment of two sentences or words. This can be used to check if two sentences are factually equivalent, or subset/superset of each other.
 
-## [`ContainsImage`](https://github.com/invariantlabs-ai/invariant/blob/main/testing/invariant/custom_types/matchers.py#L153)
+## [`ContainsImage`](https://github.com/invariantlabs-ai/testing/blob/main/invariant_testing/testing/custom_types/matchers.py#L155)
 
 Checks if the input contains an image in base64 encoding.

docs/testing/writing/parameterized-tests.md

@@ -12,7 +12,7 @@ In some cases, a certain agent functionality should generalize to multiple scena
 In `testing`, instead of writing a separate test for each city, you can use parameterized tests to test multiple scenarios. This ensures robustness and generalization of your agent's behavior.
 
 ```python
-from invariant.testing import Trace, assert_equals, parameterized
+from invariant_testing.testing import Trace, assert_equals, parameterized
 import pytest
 
 @pytest.mark.parametrize(

docs/testing/writing/tests.md

@@ -15,7 +15,7 @@ This chapter first discusses how _localized assertions_ work and then provides e
 A test case in `testing` looks a lot like a regular unit test, except that it always makes use of a `Trace` object and the corresponding `.as_context()` method. This is required to enable _localized assertions_, that maps assertions to specific ranges in the provided trace:
 
 ```python
-from invariant.testing import assert_true, Trace
+from invariant_testing.testing import assert_true, Trace
 
 def test_assert():
     # obtain some trace
@@ -44,7 +44,7 @@ def test_assert():
 To make hard (leading to test failure) assertions, you can use the `assert_true`, `assert_that`, and `assert_equals` functions. These functions are similar to the ones you might know from unit testing frameworks like `unittest` or `pytest`, but they add support for localization.
 
 ```python
-from invariant.testing import assert_true, assert_equals, assert_that, Trace, IsSimilar
+from invariant_testing.testing import assert_true, assert_equals, assert_that, Trace, IsSimilar
 
 def test_hard_assert():
     # obtain some trace
@@ -126,7 +126,7 @@ Next to hard assertions, `testing` also supports _soft assertions_ that do not l
 Instead, they are logged as warnings only and can be used to check (non-functional) agent properties that may not be critical to ensure functional correctness (e.g. number of tool calls, runtime, etc.), but are still important to monitor.
 
 ```python
-from invariant.testing import expect_equals, Trace, IsSimilar
+from invariant_testing.testing import expect_equals, Trace, IsSimilar
 
 def test_soft_assert():
     # obtain some trace

docs/testing/writing/traces.ipynb

@@ -24,7 +24,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from invariant.testing import Trace\n",
+    "from invariant_testing.testing import Trace\n",
     "\n",
     "trace = Trace(trace=[\n",
     "    {\"role\": \"user\", \"content\": \"Hello there\"},\n",