feat: add AGENTS.md

Author: adolfo-ab

.gitignore

@@ -168,7 +168,13 @@ cython_debug/
 .vscode/
 
 # AI Assistant Config Files
+.cursor/
+.clinerules/
+.roorules/
+AGENT.md
 CLAUDE.md
+GEMINI.md
+QWEN.md
 
 # Must-Gather Artifacts
 must-gather-collected/

AGENTS.md

@@ -0,0 +1,19 @@
+# Overview
+This is a testing repo for OpenDataHub and OpenShift AI, an MLOps platform for OpenShift.
+The tests contained in the repo are high-level integration tests at the Kubernetes API level.
+
+# Documentation
+All the general information about the repo is contained in the /docs directory.
+At the start of each session, consider if you need to consult any of these files in order to answer:
+- [Guidelines for Getting Started](./docs/GETTING_STARTED.md)
+- [Developer Guide](./docs/DEVELOPER_GUIDE.md)
+- [Style Guide](./docs/STYLE_GUIDE.md)
+
+# Specific Instructions
+- Avoid unnecessary complexity: Aim for the simplest solution that works, while keeping the code clean.
+- Avoid obvious comments: Only add comments to explain especially complex code blocks.
+- Maintain code consistency: Follow existing code patterns and architecture.
+- Maintain locality of behavior: Keep code close to where it's used.
+- Make small, focused changes, unless explicitly asked otherwise.
+- Keep security in mind: Avoid filtering sensitive information and running destructive commands.
+- When in doubt about something, ask the user.

docs/CONTRIBUTING.md

@@ -16,28 +16,3 @@ If a related issue doesn't exist, you can open a new issue using a relevant [iss
 
 ## Pull requests
 Follow the guidelines in [Developer guide](DEVELOPER_GUIDE.md)
-
-
-## Adding new runtime
-To add a new runtime, you need to:  
-1. Add a new file under [manifests](../utilities/manifests) directory.
-2. Add `<runtime>_INFERENCE_CONFIG` dict with:
-```code
-    "support_multi_default_queries": True|False,  # Optioanl, if set to True, `default_query_model` should contains a dict with corresponding inference_type
-    "default_query_model": {
-        "query_input": <default query to be sent to the model>,
-        "query_output": <expected output>,
-        "use_regex": True|False, # Optional, if set to True, `query_output` should be a regex
-    },
-    "<query type, for example: all-tokens>": {
-        "<protocol, for example HTTP>": {
-            "endpoint": "<model endpoint>",
-            "header": "<model required headers>",
-            "body": '{<model expected body}',
-            "response_fields_map": {
-                "response_output": <output field in response>,
-                "response": <response field in response - optional>,
-            },
-        },
-```
-3. See [caikit_standalone](../utilities/manifests/caikit_standalone.py) for an example

docs/DEVELOPER_GUIDE.md

@@ -176,3 +176,35 @@ Run tox:
 ```bash
 tox
 ```
+
+## Adding new runtime
+To add a new runtime, you need to:  
+1. Add a new file under [manifests](../utilities/manifests) directory.
+2. Add `<runtime>_INFERENCE_CONFIG` dict with:
+```code
+    "support_multi_default_queries": True|False,  # Optioanl, if set to True, `default_query_model` should contains a dict with corresponding inference_type
+    "default_query_model": {
+        "query_input": <default query to be sent to the model>,
+        "query_output": <expected output>,
+        "use_regex": True|False, # Optional, if set to True, `query_output` should be a regex
+    },
+    "<query type, for example: all-tokens>": {
+        "<protocol, for example HTTP>": {
+            "endpoint": "<model endpoint>",
+            "header": "<model required headers>",
+            "body": '{<model expected body}',
+            "response_fields_map": {
+                "response_output": <output field in response>,
+                "response": <response field in response - optional>,
+            },
+        },
+```
+3. See [caikit_standalone](../utilities/manifests/caikit_standalone.py) for an example
+
+## AI Usage
+If using AI tooling to assist you in the process of writing or reviewing code:
+1. Understand what you are doing --as a developer, you are ultimately responsible for the code. Always assume the code produced by the AI tools is unsafe and incorrect, and always double-check it.
+2. We support [AGENTS.md](../AGENTS.md), an [open format](https://agents.md/) for guiding coding agents. If you use any proprietary tool that do not support `AGENTS.md` (e.g. Claude Code, Qwen Code, Gemini Code), you can create a symlink:
+```bash
+ln -s AGENTS.md CLAUDE.md
+```