fix: Update documentation and refactor test code

- Improved documentation clarity in AGENTS.md (debugging section, reporting guidelines)
- Added log color CLI parameter documentation
- Refactored test assertions to use next() over list comprehension indexing
- Removed obsolete reproduce_issue.py test script

Author: Mateusz

AGENTS.md

@@ -61,7 +61,7 @@ When working on specs, the user will invoke `/kiro:*` commands. Follow the instr
 | **Test (Full)** | `./.venv/Scripts/python.exe -m pytest -m "integration or unit"` |
 | **Lint/Fix** | `./.venv/Scripts/python.exe -m ruff --fix check .` |
 | **Format** | `./.venv/Scripts/python.exe -m black .` |
-| **Inspect** | `./.venv/Scripts/python.exe scripts/inspect_cbor_capture.py <file> --detect-issues` |
+| **Inspect CBOR wire captures** | `./.venv/Scripts/python.exe scripts/inspect_cbor_capture.py <file> --detect-issues` |
 
 ## Quality & Testing Standards
 
@@ -82,7 +82,11 @@ When working on specs, the user will invoke `/kiro:*` commands. Follow the instr
 - **Paths**: Use `pathlib` or `/` forward slashes (Windows accepts them).
 - **Errors**: Don't use bare `except Exception`. Log with `exc_info=True`.
 
+## Debugging
+
+- **Create repro scripts**: If you cannot reproduce based on user-provided context create repro script first.
+
 ## Reporting Back To The User
 
-- **NEVER** claim you successfully fixed a bug or implemented a feature if you're not 100% sure. The only way to be sure is to run tests.
-- **ALWAYS** prove correctness by running tests.
+- **NEVER** claim you successfully fixed a bug or implemented a feature if you're not 100% sure. The only way to be sure is to run tests and/or demo script(s).
+- **ALWAYS** prove correctness by running tests before reporting back to the user.

docs/user_guide/cli-parameters.md

@@ -118,6 +118,8 @@ Configuration is resolved in the following order (highest to lowest priority):
 | :--- | :--- | :--- |
 | `--log-level LEVEL` | `LOG_LEVEL` | Logging level (DEBUG, INFO, WARNING, ERROR, CRITICAL). |
 | `--log FILE` | `LOG_FILE` | Path to log file. |
+| `--log-colors` | `LOG_COLORS=true` | Enable colored log output. |
+| `--no-log-colors` | `LOG_COLORS=false` | Disable colored log output. |
 | `--capture-file FILE` | `CAPTURE_FILE` | Write raw LLM requests/replies to this file (JSON). |
 | `--capture-max-bytes N` | `CAPTURE_MAX_BYTES` | Max size of capture file before rotation. |
 | `--capture-truncate-bytes N` | `CAPTURE_TRUNCATE_BYTES` | Truncate captures to N bytes per entry. |

tests/reproduce_issue.py

@@ -59,9 +59,9 @@ def test_extra_content_removed_from_embedded_tool_calls(self) -> None:
 
         # Parse the SSE data
         lines = result_str.strip().split("\n")
-        data_line = [
+        data_line = next(
             line for line in lines if line.startswith("data: ") and "[DONE]" not in line
-        ][0]
+        )
         json_data = json.loads(data_line[6:])  # Remove "data: " prefix
 
         # Verify extra_content is NOT in the output
@@ -100,9 +100,9 @@ def test_extra_content_removed_from_metadata_tool_calls(self) -> None:
         result_str = result.decode("utf-8")
 
         # Parse the SSE data
-        data_line = [
+        data_line = next(
             line for line in result_str.strip().split("\n") if line.startswith("data: ")
-        ][0]
+        )
         json_data = json.loads(data_line[6:])
 
         tool_calls = json_data["choices"][0]["delta"]["tool_calls"]
@@ -148,11 +148,11 @@ def test_standard_tool_call_fields_preserved(self) -> None:
         result = chunk.to_bytes()
         result_str = result.decode("utf-8")
 
-        data_line = [
+        data_line = next(
             line
             for line in result_str.strip().split("\n")
             if line.startswith("data: ") and "[DONE]" not in line
-        ][0]
+        )
         json_data = json.loads(data_line[6:])
 
         tool_calls = json_data["choices"][0]["delta"]["tool_calls"]
@@ -204,11 +204,11 @@ def test_multiple_tool_calls_all_sanitized(self) -> None:
         result = chunk.to_bytes()
         result_str = result.decode("utf-8")
 
-        data_line = [
+        data_line = next(
             line
             for line in result_str.strip().split("\n")
             if line.startswith("data: ") and "[DONE]" not in line
-        ][0]
+        )
         json_data = json.loads(data_line[6:])
 
         tool_calls = json_data["choices"][0]["delta"]["tool_calls"]