ruleprobe/action.yml at main · moonrunnerkc/ruleprobe · 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
name: "RuleProbe: Verify Instruction Adherence"
description: >
  Verify whether AI coding agent output follows project instruction file rules.
  Deterministic checks, no LLM calls, no API keys beyond GITHUB_TOKEN.

branding:
  icon: check-circle
  color: blue

inputs:
  instruction-file:
    description: "Path to instruction file (CLAUDE.md, AGENTS.md, .cursorrules, etc)"
    required: true
  output-dir:
    description: "Directory containing code to verify"
    required: true
    default: "src"
  agent:
    description: "Agent identifier for the report"
    required: false
    default: "ci"
  model:
    description: "Model identifier for the report"
    required: false
    default: "unknown"
  format:
    description: "Report format: text, json, or markdown"
    required: false
    default: "text"
  severity:
    description: "Minimum severity to report: error, warning, or all"
    required: false
    default: "all"
  fail-on-violation:
    description: "Fail the action if any rule violations are found"
    required: false
    default: "true"
  post-comment:
    description: "Post results as a PR comment"
    required: false
    default: "true"
  reviewdog-format:
    description: "Also output in reviewdog rdjson format for annotation integration"
    required: false
    default: "false"

outputs:
  score:
    description: "Adherence score as a percentage"
    value: ${{ steps.verify.outputs.score }}
  passed:
    description: "Number of rules that passed"
    value: ${{ steps.verify.outputs.passed }}
  failed:
    description: "Number of rules that failed"
    value: ${{ steps.verify.outputs.failed }}
  total:
    description: "Total number of rules checked"
    value: ${{ steps.verify.outputs.total }}

runs:
  using: "composite"
  steps:
    - name: Set up Node.js
      uses: actions/setup-node@v4
      with:
        node-version: "22"

    - name: Install ruleprobe
      shell: bash
      run: |
        cd "${{ github.action_path }}"
        npm ci --ignore-scripts
        npm run build

    - name: Run ruleprobe verify
      id: verify
      shell: bash
      env:
        INPUT_INSTRUCTION_FILE: ${{ inputs.instruction-file }}
        INPUT_OUTPUT_DIR: ${{ inputs.output-dir }}
        INPUT_AGENT: ${{ inputs.agent }}
        INPUT_MODEL: ${{ inputs.model }}
        INPUT_FORMAT: ${{ inputs.format }}
        INPUT_SEVERITY: ${{ inputs.severity }}
        INPUT_FAIL_ON_VIOLATION: ${{ inputs.fail-on-violation }}
        INPUT_REVIEWDOG_FORMAT: ${{ inputs.reviewdog-format }}
        RULEPROBE_BIN: ${{ github.action_path }}/dist/cli.js
      run: bash "${{ github.action_path }}/action-scripts/run-verify.sh"

    - name: Post PR comment
      if: inputs.post-comment == 'true' && github.event_name == 'pull_request'
      shell: bash
      env:
        GITHUB_TOKEN: ${{ github.token }}
        INPUT_FAIL_ON_VIOLATION: ${{ inputs.fail-on-violation }}
        RULEPROBE_REPORT_FILE: ${{ github.workspace }}/.ruleprobe-report.txt
      run: bash "${{ github.action_path }}/action-scripts/post-comment.sh"