Fix bug in mcp plugin #81
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ng conventions (#80) Co-authored-by: Mario Korte <mko@biss-net.com>
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
6 issues found across 4 files
Prompt for AI agents (all 6 issues)
Check if these issues are valid — if so, understand the root cause of each and fix them.
<file name="plugins/communication_protocols/mcp/tests/test_mcp_http_transport.py">
<violation number="1" location="plugins/communication_protocols/mcp/tests/test_mcp_http_transport.py:139">
P2: `transport.call_tool` returns the raw string greeting, not a `{"result": ...}` dict, so this assertion will always fail.</violation>
<violation number="2" location="plugins/communication_protocols/mcp/tests/test_mcp_http_transport.py:155">
P2: List output from `transport.call_tool` is a plain list, so asserting it is a dict with a `result` key is incorrect and breaks the test.</violation>
<violation number="3" location="plugins/communication_protocols/mcp/tests/test_mcp_http_transport.py:173">
P2: `transport.call_tool` returns the integer sum, not a dict, so this assertion is incorrect.</violation>
</file>
<file name="plugins/communication_protocols/mcp/tests/test_mcp_transport.py">
<violation number="1" location="plugins/communication_protocols/mcp/tests/test_mcp_transport.py:101">
P2: `transport.call_tool` returns the raw greeting string, but this test now expects a dict, so it will always fail.</violation>
<violation number="2" location="plugins/communication_protocols/mcp/tests/test_mcp_transport.py:111">
P2: List results are returned directly, but the test now insists on a dict wrapper, causing a deterministic failure.</violation>
<violation number="3" location="plugins/communication_protocols/mcp/tests/test_mcp_transport.py:123">
P2: Numeric tool results are returned as plain numbers, but the test now expects a dict wrapper, so it will always fail.</violation>
</file>
Reply to cubic to teach it or ask questions. Re-run a review with @cubic-dev-ai review this PR
| result = await transport.call_tool(None, f"{HTTP_SERVER_NAME}.add_numbers", {"a": 5, "b": 7}, http_mcp_provider) | ||
|
|
||
| assert result == 12 | ||
| assert result == {"result": 12} |
Contributor
There was a problem hiding this comment.
P2: transport.call_tool returns the integer sum, not a dict, so this assertion is incorrect.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At plugins/communication_protocols/mcp/tests/test_mcp_http_transport.py, line 173:
<comment>`transport.call_tool` returns the integer sum, not a dict, so this assertion is incorrect.</comment>
<file context>
@@ -172,7 +170,7 @@ async def test_http_numeric_output(
result = await transport.call_tool(None, f"{HTTP_SERVER_NAME}.add_numbers", {"a": 5, "b": 7}, http_mcp_provider)
- assert result == 12
+ assert result == {"result": 12}
</file context>
Suggested change
| assert result == {"result": 12} | |
| assert result == 12 |
✅ Addressed in 5f25643
Comment on lines
155
to
157
| assert isinstance(result, dict) | ||
| assert "result" in result | ||
| assert result == {"result": ["item_0", "item_1", "item_2"]} |
Contributor
There was a problem hiding this comment.
P2: List output from transport.call_tool is a plain list, so asserting it is a dict with a result key is incorrect and breaks the test.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At plugins/communication_protocols/mcp/tests/test_mcp_http_transport.py, line 155:
<comment>List output from `transport.call_tool` is a plain list, so asserting it is a dict with a `result` key is incorrect and breaks the test.</comment>
<file context>
@@ -152,11 +152,9 @@ async def test_http_list_output(
- assert result[0] == "item_0"
- assert result[1] == "item_1"
- assert result[2] == "item_2"
+ assert isinstance(result, dict)
+ assert "result" in result
+ assert result == {"result": ["item_0", "item_1", "item_2"]}
</file context>
Suggested change
| assert isinstance(result, dict) | |
| assert "result" in result | |
| assert result == {"result": ["item_0", "item_1", "item_2"]} | |
| assert isinstance(result, list) | |
| assert len(result) == 3 | |
| assert result[0] == "item_0" | |
| assert result[1] == "item_1" | |
| assert result[2] == "item_2" |
✅ Addressed in 5f25643
| # Call the greet tool and verify the result | ||
| result = await transport.call_tool(None, f"{HTTP_SERVER_NAME}.greet", {"name": "Alice"}, http_mcp_provider) | ||
| assert result == "Hello, Alice!" | ||
| assert result == {"result": "Hello, Alice!"} |
Contributor
There was a problem hiding this comment.
P2: transport.call_tool returns the raw string greeting, not a {"result": ...} dict, so this assertion will always fail.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At plugins/communication_protocols/mcp/tests/test_mcp_http_transport.py, line 139:
<comment>`transport.call_tool` returns the raw string greeting, not a `{"result": ...}` dict, so this assertion will always fail.</comment>
<file context>
@@ -136,7 +136,7 @@ async def test_http_unstructured_output(
# Call the greet tool and verify the result
result = await transport.call_tool(None, f"{HTTP_SERVER_NAME}.greet", {"name": "Alice"}, http_mcp_provider)
- assert result == "Hello, Alice!"
+ assert result == {"result": "Hello, Alice!"}
</file context>
Suggested change
| assert result == {"result": "Hello, Alice!"} | |
| assert result == "Hello, Alice!" |
✅ Addressed in 5f25643
| result = await transport.call_tool(None, f"{SERVER_NAME}.add_numbers", {"a": 5, "b": 7}, mcp_manual) | ||
|
|
||
| assert result == 12 | ||
| assert result == {"result": 12} |
Contributor
There was a problem hiding this comment.
P2: Numeric tool results are returned as plain numbers, but the test now expects a dict wrapper, so it will always fail.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At plugins/communication_protocols/mcp/tests/test_mcp_transport.py, line 123:
<comment>Numeric tool results are returned as plain numbers, but the test now expects a dict wrapper, so it will always fail.</comment>
<file context>
@@ -120,7 +120,7 @@ async def test_numeric_output(transport: McpCommunicationProtocol, mcp_manual: M
result = await transport.call_tool(None, f"{SERVER_NAME}.add_numbers", {"a": 5, "b": 7}, mcp_manual)
- assert result == 12
+ assert result == {"result": 12}
</file context>
Suggested change
| assert result == {"result": 12} | |
| assert result == 12 |
✅ Addressed in 5f25643
Comment on lines
111
to
113
| assert isinstance(result, dict) | ||
| assert "result" in result | ||
| assert result == {"result": ["item_0", "item_1", "item_2"]} |
Contributor
There was a problem hiding this comment.
P2: List results are returned directly, but the test now insists on a dict wrapper, causing a deterministic failure.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At plugins/communication_protocols/mcp/tests/test_mcp_transport.py, line 111:
<comment>List results are returned directly, but the test now insists on a dict wrapper, causing a deterministic failure.</comment>
<file context>
@@ -108,9 +108,9 @@ async def test_list_output(transport: McpCommunicationProtocol, mcp_manual: McpC
- assert isinstance(result, list)
- assert len(result) == 3
- assert result == ["item_0", "item_1", "item_2"]
+ assert isinstance(result, dict)
+ assert "result" in result
+ assert result == {"result": ["item_0", "item_1", "item_2"]}
</file context>
Suggested change
| assert isinstance(result, dict) | |
| assert "result" in result | |
| assert result == {"result": ["item_0", "item_1", "item_2"]} | |
| assert isinstance(result, list) | |
| assert len(result) == 3 | |
| assert result == ["item_0", "item_1", "item_2"] |
✅ Addressed in 5f25643
|
|
||
| result = await transport.call_tool(None, f"{SERVER_NAME}.greet", {"name": "Alice"}, mcp_manual) | ||
| assert result == "Hello, Alice!" | ||
| assert result == {"result": "Hello, Alice!"} |
Contributor
There was a problem hiding this comment.
P2: transport.call_tool returns the raw greeting string, but this test now expects a dict, so it will always fail.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At plugins/communication_protocols/mcp/tests/test_mcp_transport.py, line 101:
<comment>`transport.call_tool` returns the raw greeting string, but this test now expects a dict, so it will always fail.</comment>
<file context>
@@ -98,7 +98,7 @@ async def test_unstructured_string_output(transport: McpCommunicationProtocol, m
result = await transport.call_tool(None, f"{SERVER_NAME}.greet", {"name": "Alice"}, mcp_manual)
- assert result == "Hello, Alice!"
+ assert result == {"result": "Hello, Alice!"}
</file context>
Suggested change
| assert result == {"result": "Hello, Alice!"} | |
| assert result == "Hello, Alice!" |
✅ Addressed in 5f25643
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary by cubic
Fixes MCP plugin result handling and aligns Tool field names with MCP conventions, preventing misparsed outputs and making unstructured results consistent. Bumps plugin version to 1.1.2.
Bug Fixes
Migration
Written for commit 5f25643. Summary will update automatically on new commits.