prodigy 0.4.4

Turn ad-hoc Claude sessions into reproducible development pipelines with parallel AI agents
Documentation 
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

# MapReduce Workflow with Environment Variables Example
# Demonstrates environment variable usage across all workflow phases
name: mapreduce-env-example
mode: mapreduce

# Environment variables for parameterization
env:
  # Project configuration
  PROJECT_NAME: "example-project"
  PROJECT_CONFIG: "config.yml"
  FEATURES_PATH: "features"

  # Output settings
  OUTPUT_DIR: "output"
  REPORT_FORMAT: "json"

  # Workflow settings
  MAX_RETRIES: "3"
  TIMEOUT_SECONDS: "300"
  DEBUG_MODE: "false"

  # Secret variables (masked in logs)
  API_TOKEN:
    secret: true
    value: "${GITHUB_TOKEN}"

# Optional: Profile-specific configurations
profiles:
  development:
    description: "Development environment with debug enabled"
    DEBUG_MODE: "true"
    TIMEOUT_SECONDS: "60"
    OUTPUT_DIR: "dev-output"

  production:
    description: "Production environment"
    DEBUG_MODE: "false"
    TIMEOUT_SECONDS: "300"
    OUTPUT_DIR: "prod-output"

# Setup phase: Initialize with environment variables
setup:
  - shell: "echo Starting $PROJECT_NAME workflow"
  - shell: "echo Debug mode: $DEBUG_MODE"
  - shell: "mkdir -p $OUTPUT_DIR"
  - shell: "echo Created output directory: ${OUTPUT_DIR}"

  # Generate work items using env vars
  - shell: "echo '{\"items\": [{\"name\": \"item1\", \"path\": \"file1.txt\"}, {\"name\": \"item2\", \"path\": \"file2.txt\"}]}' > items.json"

# Map phase: Process items with environment variables
map:
  input: "items.json"
  json_path: "$.items[*]"

  agent_template:
    # Environment variables available in Claude commands
    - claude: "/process-item '${item.name}' --project $PROJECT_NAME --config $PROJECT_CONFIG"

    # Environment variables in shell commands
    - shell: "echo Processing ${item.name} for project $PROJECT_NAME"
    - shell: "echo Output will be saved to $OUTPUT_DIR"

    # Using timeout from env vars
    - shell: "timeout ${TIMEOUT_SECONDS}s ./process.sh ${item.path}"
      on_failure:
        - claude: "/fix-issue '${item.name}' --max-retries $MAX_RETRIES"

  max_parallel: 5

# Reduce phase: Aggregate results with environment variables
reduce:
  - shell: "echo Aggregating results for $PROJECT_NAME"
  - claude: "/summarize ${map.results} --project $PROJECT_NAME --format $REPORT_FORMAT"

  # Save results using env vars
  - shell: "cp summary.$REPORT_FORMAT $OUTPUT_DIR/${PROJECT_NAME}-summary.$REPORT_FORMAT"
  - shell: "echo Processed ${map.successful}/${map.total} items"
  - shell: "echo Results saved to ${OUTPUT_DIR}/${PROJECT_NAME}-summary.${REPORT_FORMAT}"

# Merge phase: Merge changes with environment variables
merge:
  commands:
    - shell: "echo Merging changes for $PROJECT_NAME"
    - shell: "echo Debug mode was: $DEBUG_MODE"

    # Validate merge using env vars
    - claude: "/validate-merge --branch ${merge.source_branch} --project $PROJECT_NAME"

    # Optional: Send notification using secrets
    # - shell: "curl -H 'Authorization: Bearer $API_TOKEN' -X POST https://api.github.com/repos/notify"

    - shell: "echo Merge completed for ${PROJECT_NAME}"