Skip to content

test(codex): coverage for MCP tool calls (SIG-198) #171

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
GuyMoses merged 2 commits into from
Jul 15, 2026
+92 −50
Merged

test(codex): coverage for MCP tool calls (SIG-198) #171

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2abbec0
test(codex): golden coverage for MCP tool calls (SIG-198)
GuyMoses Jul 14, 2026
80d48a6
Merge branch 'main' into sig-198-verify-add-golden-coverage-for-codex…
GuyMoses Jul 15, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 17 additions & 0 deletions internal/source/codex/golden_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -135,6 +135,23 @@ func TestGoldenSpanTree(t *testing.T) {
}
}

// MCP attribution invariant: Codex emits MCP tool calls with the same
// mcp__<server>__<tool> naming as Claude Code, so the pipeline's generic
// ExtractMCPServer/NormalizeMCPToolName must strip the prefix (tool name ->
// "echo") and attribute the server ("everything") with no Codex-specific code.
// Guards the captured mcp__everything__echo pair; fails even if the golden is
// blindly re-blessed with wrong values.
var sawMCP bool
for _, s := range canon {
if s.Attributes["dash0.gen_ai.tool.mcp_server"] == "everything" {
sawMCP = true
assert.Equal(t, "echo", s.Attributes["gen_ai.tool.name"],
"MCP tool name must be normalized (mcp__server__ prefix stripped)")
assert.Equal(t, "execute_tool", s.Attributes["gen_ai.operation.name"])
}
}
assert.True(t, sawMCP, "expected an MCP tool span with mcp_server attribution")

got := marshalGolden(t, canon)
goldenPath := filepath.Join("testdata", "golden_spans.json")

Expand Down
2 changes: 2 additions & 0 deletions internal/source/codex/testdata/captured_events.jsonl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,8 @@
{"session_id":"019f3be8-053a-78c3-9096-e9ab264c13a0","turn_id":"019f3be8-05bd-7fe1-a139-fb244f6c0f04","transcript_path":"/Users/guymoses/.codex/sessions/2026/07/07/rollout-2026-07-07T12-28-09-019f3be8-053a-78c3-9096-e9ab264c13a0.jsonl","cwd":"/private/tmp/claude-501/-Users-guymoses/7b839ad5-bbf3-4646-a79e-824ae1de5baa/scratchpad/codex-exp","hook_event_name":"PostToolUse","model":"gpt-5.5","permission_mode":"bypassPermissions","tool_name":"apply_patch","tool_input":{"command":"*** Begin Patch\n*** Add File: hello.txt\n+hi from codex\n*** End Patch\n"},"tool_response":"Exit code: 0\nWall time: 0.4 seconds\nOutput:\nSuccess. Updated the following files:\nA hello.txt\n","tool_use_id":"call_9pq2ChYkd2XGBYc3HpRZO29b"}
{"session_id":"019f3be8-053a-78c3-9096-e9ab264c13a0","turn_id":"019f3be8-05bd-7fe1-a139-fb244f6c0f04","transcript_path":"/Users/guymoses/.codex/sessions/2026/07/07/rollout-2026-07-07T12-28-09-019f3be8-053a-78c3-9096-e9ab264c13a0.jsonl","cwd":"/private/tmp/claude-501/-Users-guymoses/7b839ad5-bbf3-4646-a79e-824ae1de5baa/scratchpad/codex-exp","hook_event_name":"PreToolUse","model":"gpt-5.5","permission_mode":"bypassPermissions","tool_name":"Bash","tool_input":{"command":"wc -c hello.txt"},"tool_use_id":"call_ji8njWEPsfeWE9wGPL5RBT7o"}
{"session_id":"019f3be8-053a-78c3-9096-e9ab264c13a0","turn_id":"019f3be8-05bd-7fe1-a139-fb244f6c0f04","transcript_path":"/Users/guymoses/.codex/sessions/2026/07/07/rollout-2026-07-07T12-28-09-019f3be8-053a-78c3-9096-e9ab264c13a0.jsonl","cwd":"/private/tmp/claude-501/-Users-guymoses/7b839ad5-bbf3-4646-a79e-824ae1de5baa/scratchpad/codex-exp","hook_event_name":"PostToolUse","model":"gpt-5.5","permission_mode":"bypassPermissions","tool_name":"Bash","tool_input":{"command":"wc -c hello.txt"},"tool_response":" 14 hello.txt\n","tool_use_id":"call_ji8njWEPsfeWE9wGPL5RBT7o"}
{"session_id":"019f3be8-053a-78c3-9096-e9ab264c13a0","turn_id":"019f3be8-05bd-7fe1-a139-fb244f6c0f04","transcript_path":"/Users/guymoses/.codex/sessions/2026/07/07/rollout-2026-07-07T12-28-09-019f3be8-053a-78c3-9096-e9ab264c13a0.jsonl","cwd":"/private/tmp/claude-501/-Users-guymoses/7b839ad5-bbf3-4646-a79e-824ae1de5baa/scratchpad/codex-exp","hook_event_name":"PreToolUse","model":"gpt-5.5","permission_mode":"bypassPermissions","tool_name":"mcp__everything__echo","tool_input":{"message":"dash0-mcp-probe"},"tool_use_id":"call_ljaDC0sXe6Ys52D2uKiWf0KN"}
{"session_id":"019f3be8-053a-78c3-9096-e9ab264c13a0","turn_id":"019f3be8-05bd-7fe1-a139-fb244f6c0f04","transcript_path":"/Users/guymoses/.codex/sessions/2026/07/07/rollout-2026-07-07T12-28-09-019f3be8-053a-78c3-9096-e9ab264c13a0.jsonl","cwd":"/private/tmp/claude-501/-Users-guymoses/7b839ad5-bbf3-4646-a79e-824ae1de5baa/scratchpad/codex-exp","hook_event_name":"PostToolUse","model":"gpt-5.5","permission_mode":"bypassPermissions","tool_name":"mcp__everything__echo","tool_input":{"message":"dash0-mcp-probe"},"tool_response":{"content":[{"type":"text","text":"Echo: dash0-mcp-probe"}]},"tool_use_id":"call_ljaDC0sXe6Ys52D2uKiWf0KN"}
{"session_id":"019f3be8-053a-78c3-9096-e9ab264c13a0","turn_id":"019f3be8-05bd-7fe1-a139-fb244f6c0f04","transcript_path":"/Users/guymoses/.codex/sessions/2026/07/07/rollout-2026-07-07T12-28-09-019f3be8-053a-78c3-9096-e9ab264c13a0.jsonl","cwd":"/private/tmp/claude-501/-Users-guymoses/7b839ad5-bbf3-4646-a79e-824ae1de5baa/scratchpad/codex-exp","hook_event_name":"Stop","model":"gpt-5.5","permission_mode":"bypassPermissions","stop_hook_active":false,"last_assistant_message":"`hello.txt` is 14 bytes."}
{"session_id":"019f3bf0-9fe5-7821-b583-cd99b1eb0738","transcript_path":"/Users/guymoses/.codex/sessions/2026/07/07/rollout-2026-07-07T12-37-33-019f3bf0-9fe5-7821-b583-cd99b1eb0738.jsonl","cwd":"/private/tmp/codex-play","hook_event_name":"SessionStart","model":"gpt-5.5","permission_mode":"default","source":"startup"}
{"session_id":"019f3bf0-9fe5-7821-b583-cd99b1eb0738","turn_id":"019f3bf0-d15b-7393-a174-85a31f0194a3","transcript_path":"/Users/guymoses/.codex/sessions/2026/07/07/rollout-2026-07-07T12-37-33-019f3bf0-9fe5-7821-b583-cd99b1eb0738.jsonl","cwd":"/private/tmp/codex-play","hook_event_name":"UserPromptSubmit","model":"gpt-5.5","permission_mode":"default","prompt":"\"create foo.txt with hello, then cat it"}
Expand Down
123 changes: 73 additions & 50 deletions internal/source/codex/testdata/golden_spans.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
"kind": 1,
"trace": "trace-1",
"span": "span-1",
"parent": "span-3",
"parent": "span-4",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.name": "codex",
Expand All @@ -26,7 +26,7 @@
"kind": 1,
"trace": "trace-1",
"span": "span-2",
"parent": "span-3",
"parent": "span-4",
"durationMs": 1000,
"attributes": {
"dash0.gen_ai.tool.bash.command_family": "wc",
Expand All @@ -45,12 +45,35 @@
}
},
{
"name": "chat gpt-5.5",
"name": "execute_tool echo",
"kind": 1,
"trace": "trace-1",
"span": "span-3",
"parent": "span-4",
"durationMs": 1000,
"attributes": {
"dash0.gen_ai.tool.mcp_server": "everything",
"gen_ai.agent.name": "codex",
"gen_ai.conversation.id": "019f3be8-053a-78c3-9096-e9ab264c13a0",
"gen_ai.harness.name": "codex",
"gen_ai.operation.name": "execute_tool",
"gen_ai.provider.name": "openai",
"gen_ai.request.model": "gpt-5.5",
"gen_ai.tool.call.arguments": "{\"message\":\"dash0-mcp-probe\"}",
"gen_ai.tool.call.id": "call_ljaDC0sXe6Ys52D2uKiWf0KN",
"gen_ai.tool.call.result": "{\"content\":[{\"text\":\"Echo: dash0-mcp-probe\",\"type\":\"text\"}]}",
"gen_ai.tool.name": "echo",
"gen_ai.tool.type": "function",
"turn_id": "019f3be8-05bd-7fe1-a139-fb244f6c0f04"
}
},
{
"name": "chat gpt-5.5",
"kind": 1,
"trace": "trace-1",
"span": "span-4",
"parent": "",
"durationMs": 5000,
"durationMs": 7000,
"attributes": {
"gen_ai.agent.name": "codex",
"gen_ai.conversation.id": "019f3be8-053a-78c3-9096-e9ab264c13a0",
Expand All @@ -70,8 +93,8 @@
"name": "execute_tool apply_patch",
"kind": 1,
"trace": "trace-2",
"span": "span-4",
"parent": "span-6",
"span": "span-5",
"parent": "span-7",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.name": "codex",
Expand All @@ -92,8 +115,8 @@
"name": "execute_tool Bash",
"kind": 1,
"trace": "trace-2",
"span": "span-5",
"parent": "span-6",
"span": "span-6",
"parent": "span-7",
"durationMs": 1000,
"attributes": {
"dash0.gen_ai.tool.bash.command_family": "cat",
Expand All @@ -115,7 +138,7 @@
"name": "chat gpt-5.5",
"kind": 1,
"trace": "trace-2",
"span": "span-6",
"span": "span-7",
"parent": "",
"durationMs": 5000,
"attributes": {
Expand All @@ -137,8 +160,8 @@
"name": "execute_tool Agent",
"kind": 1,
"trace": "trace-3",
"span": "span-7",
"parent": "span-27",
"span": "span-8",
"parent": "span-28",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-5f80-7ca3-81a0-298149d46129",
Expand All @@ -160,8 +183,8 @@
"name": "execute_tool Agent",
"kind": 1,
"trace": "trace-3",
"span": "span-8",
"parent": "span-27",
"span": "span-9",
"parent": "span-28",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-605d-7393-ac4f-63f8dcc20260",
Expand All @@ -183,8 +206,8 @@
"name": "execute_tool Agent",
"kind": 1,
"trace": "trace-3",
"span": "span-9",
"parent": "span-27",
"span": "span-10",
"parent": "span-28",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-60f7-7db2-8a74-7cf0618742e6",
Expand All @@ -206,8 +229,8 @@
"name": "execute_tool apply_patch",
"kind": 1,
"trace": "trace-3",
"span": "span-10",
"parent": "span-8",
"span": "span-11",
"parent": "span-9",
"durationMs": 2000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-605d-7393-ac4f-63f8dcc20260",
Expand All @@ -229,8 +252,8 @@
"name": "execute_tool apply_patch",
"kind": 1,
"trace": "trace-3",
"span": "span-11",
"parent": "span-9",
"span": "span-12",
"parent": "span-10",
"durationMs": 2000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-60f7-7db2-8a74-7cf0618742e6",
Expand All @@ -252,8 +275,8 @@
"name": "execute_tool apply_patch",
"kind": 1,
"trace": "trace-3",
"span": "span-12",
"parent": "span-7",
"span": "span-13",
"parent": "span-8",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-5f80-7ca3-81a0-298149d46129",
Expand All @@ -275,8 +298,8 @@
"name": "invoke_agent worker",
"kind": 1,
"trace": "trace-3",
"span": "span-13",
"parent": "span-9",
"span": "span-14",
"parent": "span-10",
"durationMs": 8000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-60f7-7db2-8a74-7cf0618742e6",
Expand All @@ -298,8 +321,8 @@
"name": "execute_tool multi_agent_v1wait_agent",
"kind": 1,
"trace": "trace-3",
"span": "span-14",
"parent": "span-27",
"span": "span-15",
"parent": "span-28",
"durationMs": 8000,
"attributes": {
"gen_ai.agent.name": "codex",
Expand All @@ -320,8 +343,8 @@
"name": "execute_tool Bash",
"kind": 1,
"trace": "trace-3",
"span": "span-15",
"parent": "span-8",
"span": "span-16",
"parent": "span-9",
"durationMs": 1000,
"attributes": {
"dash0.gen_ai.tool.bash.command_family": "wc",
Expand All @@ -344,8 +367,8 @@
"name": "execute_tool Bash",
"kind": 1,
"trace": "trace-3",
"span": "span-16",
"parent": "span-7",
"span": "span-17",
"parent": "span-8",
"durationMs": 1000,
"attributes": {
"dash0.gen_ai.tool.bash.command_family": "od",
Expand All @@ -368,8 +391,8 @@
"name": "invoke_agent worker",
"kind": 1,
"trace": "trace-3",
"span": "span-17",
"parent": "span-8",
"span": "span-18",
"parent": "span-9",
"durationMs": 15000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-605d-7393-ac4f-63f8dcc20260",
Expand All @@ -391,8 +414,8 @@
"name": "execute_tool multi_agent_v1wait_agent",
"kind": 1,
"trace": "trace-3",
"span": "span-18",
"parent": "span-27",
"span": "span-19",
"parent": "span-28",
"durationMs": 2000,
"attributes": {
"gen_ai.agent.name": "codex",
Expand All @@ -413,8 +436,8 @@
"name": "invoke_agent worker",
"kind": 1,
"trace": "trace-3",
"span": "span-19",
"parent": "span-7",
"span": "span-20",
"parent": "span-8",
"durationMs": 17000,
"attributes": {
"gen_ai.agent.id": "019f3bf3-5f80-7ca3-81a0-298149d46129",
Expand All @@ -436,8 +459,8 @@
"name": "execute_tool multi_agent_v1wait_agent",
"kind": 1,
"trace": "trace-3",
"span": "span-20",
"parent": "span-27",
"span": "span-21",
"parent": "span-28",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.name": "codex",
Expand All @@ -458,8 +481,8 @@
"name": "execute_tool Bash",
"kind": 1,
"trace": "trace-3",
"span": "span-21",
"parent": "span-27",
"span": "span-22",
"parent": "span-28",
"durationMs": 1000,
"attributes": {
"dash0.gen_ai.tool.bash.command_family": "od",
Expand All @@ -481,8 +504,8 @@
"name": "execute_tool Bash",
"kind": 1,
"trace": "trace-3",
"span": "span-22",
"parent": "span-27",
"span": "span-23",
"parent": "span-28",
"durationMs": 2000,
"attributes": {
"dash0.gen_ai.tool.bash.command_family": "od",
Expand All @@ -504,8 +527,8 @@
"name": "execute_tool Bash",
"kind": 1,
"trace": "trace-3",
"span": "span-23",
"parent": "span-27",
"span": "span-24",
"parent": "span-28",
"durationMs": 2000,
"attributes": {
"dash0.gen_ai.tool.bash.command_family": "od",
Expand All @@ -527,8 +550,8 @@
"name": "execute_tool multi_agent_v1close_agent",
"kind": 1,
"trace": "trace-3",
"span": "span-24",
"parent": "span-27",
"span": "span-25",
"parent": "span-28",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.name": "codex",
Expand All @@ -549,8 +572,8 @@
"name": "execute_tool multi_agent_v1close_agent",
"kind": 1,
"trace": "trace-3",
"span": "span-25",
"parent": "span-27",
"span": "span-26",
"parent": "span-28",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.name": "codex",
Expand All @@ -571,8 +594,8 @@
"name": "execute_tool multi_agent_v1close_agent",
"kind": 1,
"trace": "trace-3",
"span": "span-26",
"parent": "span-27",
"span": "span-27",
"parent": "span-28",
"durationMs": 1000,
"attributes": {
"gen_ai.agent.name": "codex",
Expand All @@ -593,7 +616,7 @@
"name": "chat gpt-5.5",
"kind": 1,
"trace": "trace-3",
"span": "span-27",
"span": "span-28",
"parent": "",
"durationMs": 32000,
"attributes": {
Expand Down
Loading