playwright/action.yml at main · autospec-ai/playwright · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
name: 'AutoSpec AI - E2E Test Generator'
description: 'Automatically generates Playwright E2E tests from commit diffs using LLMs (Anthropic, OpenAI, or compatible providers)'
author: 'autospec-ai'

branding:
  icon: 'shield'
  color: 'purple'

inputs:
  # --- LLM Configuration ---
  llm_provider:
    description: 'LLM provider: anthropic | openai | custom'
    required: false
    default: 'anthropic'
  llm_api_key:
    description: 'API key for the LLM provider (use GitHub secrets)'
    required: true
  llm_model:
    description: 'Model to use (defaults: claude-sonnet-4-20250514 for anthropic, gpt-4o for openai)'
    required: false
    default: ''
  llm_base_url:
    description: 'Custom base URL for OpenAI-compatible providers (e.g., Ollama, Together, Groq)'
    required: false
    default: ''

  # --- Test Configuration ---
  test_directory:
    description: 'Directory where generated tests will be written'
    required: false
    default: 'e2e/generated'
  test_pattern:
    description: 'Comma-separated glob patterns to find existing tests (e.g., e2e/**/*.spec.ts,*-e2e/**/*.spec.ts)'
    required: false
    default: 'e2e/**/*.spec.ts,*-e2e/**/*.spec.ts'
  base_url:
    description: 'Application base URL for Playwright tests'
    required: false
    default: 'http://localhost:3000'
  framework:
    description: 'Frontend framework hint: react | vue | svelte | angular | nextjs | generic'
    required: false
    default: 'generic'

  # --- Diff Configuration ---
  diff_mode:
    description: 'How to determine changes: auto | pr | push'
    required: false
    default: 'auto'
  include_paths:
    description: 'Comma-separated path prefixes to include (e.g., src/,pages/,components/)'
    required: false
    default: ''
  exclude_paths:
    description: 'Comma-separated path prefixes to exclude (e.g., test/,docs/,.github/)'
    required: false
    default: 'test/,tests/,e2e/,__tests__/,.github/,docs/,README'

  # --- Behavior ---
  auto_commit:
    description: 'Automatically commit generated tests'
    required: false
    default: 'false'
  auto_pr:
    description: 'Create a PR with generated tests instead of committing to branch'
    required: false
    default: 'true'
  max_test_files:
    description: 'Maximum number of test files to generate per run'
    required: false
    default: '5'
  dry_run:
    description: 'Preview generated tests without writing files'
    required: false
    default: 'false'
  overwrite_existing_files:
    description: 'Allow generated tests, fixtures, or page objects to overwrite existing files'
    required: false
    default: 'false'
  custom_instructions:
    description: 'Additional instructions for the LLM when generating tests'
    required: false
    default: ''

  # --- Project Structure Discovery ---
  pom_patterns:
    description: 'Comma-separated glob patterns for page object / POM files (e.g., **/pages/**/*.ts,**/*.page.ts). Auto-detected when empty.'
    required: false
    default: ''
  utility_patterns:
    description: 'Comma-separated glob patterns for helper/utility files (e.g., **/helpers/**/*.ts,**/utils/**/*.ts). Auto-detected when empty.'
    required: false
    default: ''
  pom_output_directory:
    description: 'Directory where generated page object files will be written. When set, the LLM creates POM files separately instead of inlining classes in test specs.'
    required: false
    default: ''
  project_context_budget:
    description: 'Approximate token budget for project context injected into LLM prompts'
    required: false
    default: '8000'
  diff_context_budget:
    description: 'Approximate token budget for changed-file diffs and source injected into planning prompts'
    required: false
    default: '24000'

  # --- Trace Viewer Integration ---
  trace_on_failure:
    description: 'Enable Playwright trace collection for test failures'
    required: false
    default: 'true'
  trace_mode:
    description: 'Playwright trace mode: on | off | retain-on-failure | on-first-retry'
    required: false
    default: 'retain-on-failure'
  test_results_directory:
    description: 'Directory scanned by the post-action for Playwright traces, screenshots, and video'
    required: false
    default: 'test-results'

  # --- API Mock Generation ---
  generate_api_mocks:
    description: 'Detect API dependencies and generate page.route() mocks in tests'
    required: false
    default: 'false'
  mock_error_states:
    description: 'Generate additional test cases for API error responses (4xx/5xx)'
    required: false
    default: 'false'
  fixture_extraction_threshold:
    description: 'Number of page.route() calls before extracting mocks into a shared fixture file'
    required: false
    default: '3'

  # --- Visual Regression Baselines ---
  visual_regression:
    description: 'Add toHaveScreenshot() assertions at visual checkpoints'
    required: false
    default: 'false'
  visual_threshold:
    description: 'Pixel comparison threshold (0-1) for visual regression'
    required: false
    default: '0.2'
  visual_max_diff_ratio:
    description: 'Maximum allowed diff pixel ratio (0-1) before test fails'
    required: false
    default: '0.05'
  visual_full_page:
    description: 'Capture full-page screenshots instead of viewport only'
    required: false
    default: 'false'

  # --- Accessibility / Aria Snapshot Assertions ---
  accessibility_assertions:
    description: 'Add toMatchAriaSnapshot() assertions for changed components'
    required: false
    default: 'false'
  axe_scan:
    description: 'Generate a dedicated axe-core accessibility scan test case'
    required: false
    default: 'false'
  axe_standard:
    description: 'axe-core accessibility standard: wcag2a | wcag2aa | wcag21a | wcag21aa | best-practice'
    required: false
    default: 'wcag2aa'

outputs:
  tests_generated:
    description: 'Number of test files generated'
  test_files:
    description: 'JSON array of generated test file paths'
  pr_number:
    description: 'PR number if auto_pr is enabled'
  summary:
    description: 'Human-readable summary of what was generated'
  fixture_files:
    description: 'JSON array of generated fixture file paths (when API mock generation is enabled)'
  pom_files:
    description: 'JSON array of generated page-object file paths'

runs:
  using: 'node24'
  main: 'dist/index.mjs'
  post: 'dist/post/index.mjs'
  post-if: 'always()'