qaa-agent 1.9.2 → 1.9.5
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +206 -179
- package/CLAUDE.md +718 -557
- package/README.md +384 -357
- package/VERSION +1 -1
- package/agents/qa-pipeline-orchestrator.md +1739 -1425
- package/agents/qaa-analyzer.md +0 -1
- package/agents/qaa-bug-detective.md +790 -655
- package/agents/qaa-codebase-mapper.md +50 -1
- package/agents/qaa-discovery.md +421 -384
- package/agents/qaa-e2e-runner.md +715 -577
- package/agents/qaa-executor.md +947 -830
- package/agents/qaa-planner.md +14 -1
- package/agents/qaa-project-researcher.md +533 -400
- package/agents/qaa-scanner.md +77 -1
- package/agents/qaa-testid-injector.md +0 -1
- package/agents/qaa-validator.md +86 -1
- package/bin/install.cjs +375 -253
- package/bin/lib/context7-cache.cjs +299 -0
- package/bin/lib/intent-detector.cjs +488 -0
- package/commands/qa-audit.md +255 -126
- package/commands/qa-create-test.md +666 -365
- package/commands/qa-fix.md +684 -513
- package/commands/qa-map.md +283 -139
- package/commands/qa-pr.md +63 -0
- package/commands/qa-research.md +181 -157
- package/commands/qa-start.md +62 -6
- package/commands/qa-test-report.md +219 -219
- package/frameworks/cypress.json +54 -0
- package/frameworks/jest.json +59 -0
- package/frameworks/playwright.json +58 -0
- package/frameworks/pytest.json +59 -0
- package/frameworks/robot-framework.json +54 -0
- package/frameworks/selenium.json +65 -0
- package/frameworks/vitest.json +57 -0
- package/package.json +5 -2
- package/workflows/qa-analyze.md +100 -4
- package/workflows/qa-from-ticket.md +50 -2
- package/workflows/qa-gap.md +50 -2
- package/workflows/qa-start.md +2048 -1405
- package/workflows/qa-testid.md +50 -2
- package/workflows/qa-validate.md +50 -2
package/workflows/qa-testid.md
CHANGED
|
@@ -132,8 +132,32 @@ Spawn the scanner agent to produce SCAN_MANIFEST.md with the frontend component
|
|
|
132
132
|
```
|
|
133
133
|
Task(
|
|
134
134
|
prompt="
|
|
135
|
+
<critical_directive priority="MANDATORY">
|
|
136
|
+
You are executing as a QAA pipeline agent. Your FIRST action MUST be to read
|
|
137
|
+
~/.claude/qaa/agents/qaa-scanner.md and adopt it as your operating contract.
|
|
138
|
+
|
|
139
|
+
Even though you may be running as a general-purpose agent (native subagent type
|
|
140
|
+
routing for QAA agents is not yet enabled), you MUST behave exactly as the
|
|
141
|
+
dedicated agent defined in that file: read everything in its <required_reading>,
|
|
142
|
+
obey its <quality_gate>, and run its mandatory bash checklist before returning.
|
|
143
|
+
|
|
144
|
+
Never modify application source code (src/, app/, lib/, or equivalent
|
|
145
|
+
directories) unless the agent's <scope> section explicitly permits it. If the
|
|
146
|
+
agent's contract restricts what files it can touch, honor that restriction —
|
|
147
|
+
even if a fix seems obvious or trivial.
|
|
148
|
+
|
|
149
|
+
If you cannot read ~/.claude/qaa/agents/qaa-scanner.md (file missing, corrupted, parse
|
|
150
|
+
error), STOP and report the failure. Do NOT proceed with default agent
|
|
151
|
+
behavior — that defeats the purpose of this directive.
|
|
152
|
+
|
|
153
|
+
Do NOT improvise, substitute steps, apply changes the agent isn't allowed to
|
|
154
|
+
make, or skip any of the above. Skipping the agent's required_reading,
|
|
155
|
+
quality_gate, or bash checklist makes the run INVALID — its output cannot be
|
|
156
|
+
trusted.
|
|
157
|
+
</critical_directive>
|
|
158
|
+
|
|
135
159
|
<objective>Scan frontend repository and produce SCAN_MANIFEST.md with component file enumeration</objective>
|
|
136
|
-
<execution_context
|
|
160
|
+
<execution_context>~/.claude/qaa/agents/qaa-scanner.md</execution_context>
|
|
137
161
|
<files_to_read>
|
|
138
162
|
- CLAUDE.md
|
|
139
163
|
</files_to_read>
|
|
@@ -180,8 +204,32 @@ Spawn the testid-injector agent which operates in 4 phases: SCAN, AUDIT, INJECT,
|
|
|
180
204
|
```
|
|
181
205
|
Task(
|
|
182
206
|
prompt="
|
|
207
|
+
<critical_directive priority="MANDATORY">
|
|
208
|
+
You are executing as a QAA pipeline agent. Your FIRST action MUST be to read
|
|
209
|
+
~/.claude/qaa/agents/qaa-testid-injector.md and adopt it as your operating contract.
|
|
210
|
+
|
|
211
|
+
Even though you may be running as a general-purpose agent (native subagent type
|
|
212
|
+
routing for QAA agents is not yet enabled), you MUST behave exactly as the
|
|
213
|
+
dedicated agent defined in that file: read everything in its <required_reading>,
|
|
214
|
+
obey its <quality_gate>, and run its mandatory bash checklist before returning.
|
|
215
|
+
|
|
216
|
+
Never modify application source code (src/, app/, lib/, or equivalent
|
|
217
|
+
directories) unless the agent's <scope> section explicitly permits it. If the
|
|
218
|
+
agent's contract restricts what files it can touch, honor that restriction —
|
|
219
|
+
even if a fix seems obvious or trivial.
|
|
220
|
+
|
|
221
|
+
If you cannot read ~/.claude/qaa/agents/qaa-testid-injector.md (file missing, corrupted, parse
|
|
222
|
+
error), STOP and report the failure. Do NOT proceed with default agent
|
|
223
|
+
behavior — that defeats the purpose of this directive.
|
|
224
|
+
|
|
225
|
+
Do NOT improvise, substitute steps, apply changes the agent isn't allowed to
|
|
226
|
+
make, or skip any of the above. Skipping the agent's required_reading,
|
|
227
|
+
quality_gate, or bash checklist makes the run INVALID — its output cannot be
|
|
228
|
+
trusted.
|
|
229
|
+
</critical_directive>
|
|
230
|
+
|
|
183
231
|
<objective>Audit frontend components for data-testid coverage and inject missing attributes. Use Playwright MCP to verify against live DOM. Use codebase map for component structure context. Update locator registry with discovered and injected locators.</objective>
|
|
184
|
-
<execution_context
|
|
232
|
+
<execution_context>~/.claude/qaa/agents/qaa-testid-injector.md</execution_context>
|
|
185
233
|
<files_to_read>
|
|
186
234
|
- {OUTPUT_DIR}/SCAN_MANIFEST.md
|
|
187
235
|
- CLAUDE.md
|
package/workflows/qa-validate.md
CHANGED
|
@@ -131,8 +131,32 @@ Spawn the validator agent to perform 4-layer validation with fix loop.
|
|
|
131
131
|
```
|
|
132
132
|
Task(
|
|
133
133
|
prompt="
|
|
134
|
+
<critical_directive priority="MANDATORY">
|
|
135
|
+
You are executing as a QAA pipeline agent. Your FIRST action MUST be to read
|
|
136
|
+
~/.claude/qaa/agents/qaa-validator.md and adopt it as your operating contract.
|
|
137
|
+
|
|
138
|
+
Even though you may be running as a general-purpose agent (native subagent type
|
|
139
|
+
routing for QAA agents is not yet enabled), you MUST behave exactly as the
|
|
140
|
+
dedicated agent defined in that file: read everything in its <required_reading>,
|
|
141
|
+
obey its <quality_gate>, and run its mandatory bash checklist before returning.
|
|
142
|
+
|
|
143
|
+
Never modify application source code (src/, app/, lib/, or equivalent
|
|
144
|
+
directories) unless the agent's <scope> section explicitly permits it. If the
|
|
145
|
+
agent's contract restricts what files it can touch, honor that restriction —
|
|
146
|
+
even if a fix seems obvious or trivial.
|
|
147
|
+
|
|
148
|
+
If you cannot read ~/.claude/qaa/agents/qaa-validator.md (file missing, corrupted, parse
|
|
149
|
+
error), STOP and report the failure. Do NOT proceed with default agent
|
|
150
|
+
behavior — that defeats the purpose of this directive.
|
|
151
|
+
|
|
152
|
+
Do NOT improvise, substitute steps, apply changes the agent isn't allowed to
|
|
153
|
+
make, or skip any of the above. Skipping the agent's required_reading,
|
|
154
|
+
quality_gate, or bash checklist makes the run INVALID — its output cannot be
|
|
155
|
+
trusted.
|
|
156
|
+
</critical_directive>
|
|
157
|
+
|
|
134
158
|
<objective>Validate test files against CLAUDE.md standards across 4 layers (Syntax, Structure, Dependencies, Logic). Use codebase map for context and locator registry to verify locator accuracy.</objective>
|
|
135
|
-
<execution_context
|
|
159
|
+
<execution_context>~/.claude/qaa/agents/qaa-validator.md</execution_context>
|
|
136
160
|
<files_to_read>
|
|
137
161
|
- CLAUDE.md
|
|
138
162
|
- {synthetic file list or generation plan path}
|
|
@@ -191,8 +215,32 @@ if unresolved_count is 0:
|
|
|
191
215
|
```
|
|
192
216
|
Task(
|
|
193
217
|
prompt="
|
|
218
|
+
<critical_directive priority="MANDATORY">
|
|
219
|
+
You are executing as a QAA pipeline agent. Your FIRST action MUST be to read
|
|
220
|
+
~/.claude/qaa/agents/qaa-bug-detective.md and adopt it as your operating contract.
|
|
221
|
+
|
|
222
|
+
Even though you may be running as a general-purpose agent (native subagent type
|
|
223
|
+
routing for QAA agents is not yet enabled), you MUST behave exactly as the
|
|
224
|
+
dedicated agent defined in that file: read everything in its <required_reading>,
|
|
225
|
+
obey its <quality_gate>, and run its mandatory bash checklist before returning.
|
|
226
|
+
|
|
227
|
+
Never modify application source code (src/, app/, lib/, or equivalent
|
|
228
|
+
directories) unless the agent's <scope> section explicitly permits it. If the
|
|
229
|
+
agent's contract restricts what files it can touch, honor that restriction —
|
|
230
|
+
even if a fix seems obvious or trivial.
|
|
231
|
+
|
|
232
|
+
If you cannot read ~/.claude/qaa/agents/qaa-bug-detective.md (file missing, corrupted, parse
|
|
233
|
+
error), STOP and report the failure. Do NOT proceed with default agent
|
|
234
|
+
behavior — that defeats the purpose of this directive.
|
|
235
|
+
|
|
236
|
+
Do NOT improvise, substitute steps, apply changes the agent isn't allowed to
|
|
237
|
+
make, or skip any of the above. Skipping the agent's required_reading,
|
|
238
|
+
quality_gate, or bash checklist makes the run INVALID — its output cannot be
|
|
239
|
+
trusted.
|
|
240
|
+
</critical_directive>
|
|
241
|
+
|
|
194
242
|
<objective>Run test suite and classify failures into APPLICATION BUG, TEST CODE ERROR, ENVIRONMENT ISSUE, or INCONCLUSIVE</objective>
|
|
195
|
-
<execution_context
|
|
243
|
+
<execution_context>~/.claude/qaa/agents/qaa-bug-detective.md</execution_context>
|
|
196
244
|
<files_to_read>
|
|
197
245
|
- CLAUDE.md
|
|
198
246
|
- {test file paths with failures}
|