fix(qa-lab): hard gate runtime tool coverage

This commit is contained in:
Vincent Koc
2026-05-18 09:42:10 +08:00
parent 73f4657869
commit 58e1351863
19 changed files with 318 additions and 41 deletions

View File

@@ -28,7 +28,10 @@ Coverage tracking:
Runtime parity tiers:
- `standard`: required Codex-vs-Pi mock gate coverage for first-hour depth and
default runtime-tool fixtures; selected with
default runtime-tool fixtures. OpenClaw dynamic integration tools in this
tier are hard-gated by `openclaw qa coverage --tools --summary`; Codex-native
workspace rows remain separately tracked until native/live behavior is the
asserted surface. Selected with
`openclaw qa suite --runtime-pair pi,codex --runtime-parity-tier standard`
- `optional`: profile-, plugin-, or external-service-dependent runtime-tool
fixtures that stay out of the default release gate

View File

@@ -13,6 +13,7 @@ successCriteria:
- Effective tools expose image_generate after QA image-generation config is applied.
- The mock provider plans exactly one happy-path image_generate call.
- The mock provider plans one denied-input failure-path image_generate call.
- Runtime parity coverage hard-fails call/result drift in the standard direct-loading gate.
docsRefs:
- docs/tools/image-generation.md
codeRefs:
@@ -29,15 +30,12 @@ execution:
actualTool: image_generate
bucket: openclaw-dynamic-integration
expectedLayer: openclaw-dynamic
capabilityLayer: openclaw-dynamic-direct
required: true
tracking: "#80319"
codexDefaultImpact: P4
qaImpact: P1
action: teach fixture/mock planner Codex searchable OpenClaw dynamic tool behavior
reason: image_generate is an OpenClaw integration tool; QA mock provider does not yet model Codex searchable/deferred dynamic tool declarations for this fixture.
knownHarnessGap:
issue: "#80319"
reason: QA mock provider does not yet model Codex searchable/deferred OpenClaw dynamic tool declarations for this fixture.
action: hard gate in the standard direct-loading tier
reason: image_generate is an OpenClaw integration tool and must stay visible and callable under Pi and Codex direct runtime parity.
promptSnippet: "target=image_generate"
failurePromptSnippet: "failure target=image_generate"
```

View File

@@ -13,6 +13,7 @@ successCriteria:
- Effective tools expose session_status.
- The mock provider plans exactly one happy-path session_status call.
- The mock provider plans one denied-input failure-path session_status call.
- Runtime parity coverage hard-fails call/result drift in the standard direct-loading gate.
docsRefs:
- qa/scenarios/index.md
codeRefs:
@@ -28,15 +29,12 @@ execution:
actualTool: session_status
bucket: openclaw-dynamic-integration
expectedLayer: openclaw-dynamic
capabilityLayer: openclaw-dynamic-direct
required: true
tracking: "#80319"
codexDefaultImpact: P4
qaImpact: P1
action: teach fixture/mock planner Codex searchable OpenClaw dynamic tool behavior
reason: session_status is an OpenClaw integration tool; QA mock provider does not yet model Codex searchable/deferred dynamic tool declarations for this fixture.
knownHarnessGap:
issue: "#80319"
reason: QA mock provider does not yet model Codex searchable/deferred OpenClaw dynamic tool declarations for this fixture.
action: hard gate in the standard direct-loading tier
reason: session_status is an OpenClaw integration tool and must stay visible and callable under Pi and Codex direct runtime parity.
promptSnippet: "target=session_status"
failurePromptSnippet: "failure target=session_status"
```

View File

@@ -13,6 +13,7 @@ successCriteria:
- Effective tools expose sessions_spawn.
- The mock provider plans exactly one happy-path sessions_spawn call.
- The mock provider plans one denied-input failure-path sessions_spawn call.
- Runtime parity coverage hard-fails call/result drift in the standard direct-loading gate.
docsRefs:
- qa/scenarios/index.md
codeRefs:
@@ -28,15 +29,12 @@ execution:
actualTool: sessions_spawn
bucket: openclaw-dynamic-integration
expectedLayer: openclaw-dynamic
capabilityLayer: openclaw-dynamic-direct
required: true
tracking: "#80319"
codexDefaultImpact: P4
qaImpact: P1
action: teach fixture/mock planner Codex searchable OpenClaw dynamic tool behavior
reason: sessions_spawn is an OpenClaw integration tool; QA mock provider does not yet model Codex searchable/deferred dynamic tool declarations for this fixture.
knownHarnessGap:
issue: "#80319"
reason: QA mock provider does not yet model Codex searchable/deferred OpenClaw dynamic tool declarations for this fixture.
action: hard gate in the standard direct-loading tier
reason: sessions_spawn is an OpenClaw integration tool and must stay visible and callable under Pi and Codex direct runtime parity.
promptSnippet: "target=sessions_spawn"
failurePromptSnippet: "failure target=sessions_spawn"
```

View File

@@ -13,6 +13,7 @@ successCriteria:
- Effective tools expose web_fetch.
- The mock provider plans exactly one happy-path web_fetch call.
- The mock provider plans one denied-input failure-path web_fetch call.
- Runtime parity coverage hard-fails call/result drift in the standard direct-loading gate.
docsRefs:
- qa/scenarios/index.md
codeRefs:
@@ -28,15 +29,12 @@ execution:
actualTool: web_fetch
bucket: openclaw-dynamic-integration
expectedLayer: openclaw-dynamic
capabilityLayer: openclaw-dynamic-direct
required: true
tracking: "#80319"
codexDefaultImpact: P4
qaImpact: P1
action: teach fixture/mock planner Codex searchable OpenClaw dynamic tool behavior
reason: web_fetch is an OpenClaw integration tool; QA mock provider does not yet model Codex searchable/deferred dynamic tool declarations for this fixture.
knownHarnessGap:
issue: "#80319"
reason: QA mock provider does not yet model Codex searchable/deferred OpenClaw dynamic tool declarations for this fixture.
action: hard gate in the standard direct-loading tier
reason: web_fetch is an OpenClaw integration tool and must stay visible and callable under Pi and Codex direct runtime parity.
promptSnippet: "target=web_fetch"
failurePromptSnippet: "failure target=web_fetch"
```

View File

@@ -13,6 +13,7 @@ successCriteria:
- Effective tools expose web_search.
- The mock provider plans exactly one happy-path web_search call.
- The mock provider plans one denied-input failure-path web_search call.
- Runtime parity coverage hard-fails call/result drift in the standard direct-loading gate.
docsRefs:
- qa/scenarios/index.md
codeRefs:
@@ -28,15 +29,12 @@ execution:
actualTool: web_search
bucket: openclaw-dynamic-integration
expectedLayer: openclaw-dynamic
capabilityLayer: openclaw-dynamic-direct
required: true
tracking: "#80319"
codexDefaultImpact: P4
qaImpact: P1
action: teach fixture/mock planner Codex searchable OpenClaw dynamic tool behavior
reason: web_search is an OpenClaw integration tool; QA mock provider does not yet model Codex searchable/deferred dynamic tool declarations for this fixture.
knownHarnessGap:
issue: "#80319"
reason: QA mock provider does not yet model Codex searchable/deferred OpenClaw dynamic tool declarations for this fixture.
action: hard gate in the standard direct-loading tier
reason: web_search is an OpenClaw integration tool and must stay visible and callable under Pi and Codex direct runtime parity.
promptSnippet: "target=web_search"
failurePromptSnippet: "failure target=web_search"
```