eval-view/examples/github-workflow-example.yml at main · hidai25/eval-view · 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
# =============================================================================
# QUICK START — Copy this file to .github/workflows/evalview.yml in your repo
# =============================================================================
# Catches agent regressions on every PR. Automatically:
#   - Runs evalview check against your golden baselines
#   - Posts a PR comment with pass/fail, tool diffs, cost/latency alerts
#   - Writes a job summary visible in the Actions UI
#   - Uploads results + HTML report as artifacts
#
# Setup (3 steps):
#   1. Copy this file to .github/workflows/evalview.yml
#   2. Add OPENAI_API_KEY to repository secrets (Settings > Secrets)
#   3. Commit your .evalview/golden/ directory to git
#
# That's it. Every PR now gets regression-checked.
# =============================================================================

name: EvalView Agent Check

on:
  pull_request:
  push:
    branches: [main]

jobs:
  agent-check:
    runs-on: ubuntu-latest
    permissions:
      pull-requests: write  # Required for PR comments
    steps:
      - uses: actions/checkout@v4

      - name: Check for agent regressions
        uses: hidai25/eval-view@main
        with:
          openai-api-key: ${{ secrets.OPENAI_API_KEY }}

# =============================================================================
# COMMON CONFIGURATIONS
# =============================================================================
#
# Strict mode — fail on ANY change (tools, output, or regression):
#
#       - name: Check for agent regressions
#         uses: hidai25/eval-view@main
#         with:
#           openai-api-key: ${{ secrets.OPENAI_API_KEY }}
#           strict: 'true'
#
# Budget cap — limit LLM spend per CI run:
#
#       - name: Check for agent regressions
#         uses: hidai25/eval-view@main
#         with:
#           openai-api-key: ${{ secrets.OPENAI_API_KEY }}
#           budget: '1.00'
#
# Check only specific tests:
#
#       - name: Check for agent regressions
#         uses: hidai25/eval-view@main
#         with:
#           openai-api-key: ${{ secrets.OPENAI_API_KEY }}
#           filter: 'customer-support'
#
# Full evaluation mode (scores + judge, not just regression diff):
#
#       - name: Run full evaluation
#         uses: hidai25/eval-view@main
#         with:
#           openai-api-key: ${{ secrets.OPENAI_API_KEY }}
#           mode: 'run'
#
# Pin EvalView version:
#
#       - name: Check for agent regressions
#         uses: hidai25/eval-view@main
#         with:
#           openai-api-key: ${{ secrets.OPENAI_API_KEY }}
#           evalview-version: '0.5.3'
#
# Skip PR comments (CI-only, no noise):
#
#       - name: Check for agent regressions
#         uses: hidai25/eval-view@main
#         with:
#           openai-api-key: ${{ secrets.OPENAI_API_KEY }}
#           post-comment: 'false'