Analysis Orchestrator

Data Analysis Orchestrator

Core Principles

Use data to reduce guesswork in decisions

The value of data analysis lies not in producing reports, but in converting uncertainty into quantifiable risk and transforming intuitive judgments into evidence-backed decisions.

Orchestration Philosophy

1. Detection first, analysis follows, report wraps up: Anomaly detection runs 24/7, funnel and retention triggered on demand, report integrates and closes
2. Every analysis result must be actionable: Analysis results without action recommendations are not passed downstream
3. Anomalies block, others run in sequence: P0 anomalies immediately block current flow, other stages run in sequence

Orchestration Protocol

The orchestration protocol follows the unified standard in [orchestrator-protocol.md](../../../codex-templates/orchestrator-protocol.md).

Pipeline

pipeline: analysis-orchestrator
version: 7.1
post_pipeline:
  - action: stage-summary
    output: output/phase-reports/pm-metrics-ops/analysis-orchestrator.md

stages:
  - id: phase-1
    name: "Anomaly Detection"
    depends_on: []
    skills: [analysis-anomaly]
    gate:
      condition: "Anomaly detection pipeline running continuously, no interruptions"
      fail_action: "Immediately fix detection pipeline, activate backup monitoring"

  - id: phase-2
    name: "Funnel Analysis"
    parallel_with: [phase-3]
    skills: [analysis-funnel]
    gate:
      condition: "Core business funnel defined and data complete"
      fail_action: "Supplement funnel definition, ensure core path coverage"

  - id: phase-3
    name: "Retention Analysis"
    parallel_with: [phase-2]
    skills: [analysis-retention]
    gate:
      condition: "At least 1 Aha Moment candidate behavior produced"
      fail_action: "Expand behavior search scope or extend analysis period"

  - id: phase-4
    name: "Data Analysis Report"
    depends_on: [phase-1, phase-2, phase-3]
    skills: [data-analysis-report]
    gate:
      condition: "Report executive summary complete, at least 3 action recommendations"
      fail_action: "Supplement analysis or mark recommendations as needing additional data"

Stage Execution Plan

Invoke analysis-anomaly

Invoke: ${analysis-anomaly}
Input:
  metrics_system: metrics-system -> metrics.json
  real_time_data: data warehouse / real-time compute platform
  alert_rules: provided by user
  event_calendar: provided by user (optional)
Output: output/pm-metrics-ops/analysis-anomaly/
Validation: Anomaly detection covers all key metrics; anomaly severity classified correctly (P0/P1/P2); root cause analysis supported by data; recommended actions are actionable
Mode: AI

Invoke analysis-funnel

Invoke: ${analysis-funnel}
Input:
  funnel_definition: provided by user
  event_data: provided by user
  segment_config: provided by user (optional)
  comparison_period: provided by user (optional)
Output: output/pm-metrics-ops/analysis-funnel/
Validation: Funnel steps defined completely without omissions; conversion rates calculated from full data; drop-off nodes identified with cause hypotheses; multi-dimensional drill-down covers at least 3 dimensions
Mode: AI

Invoke analysis-retention

Invoke: ${analysis-retention}
Input:
  user_behavior_data: provided by user
  segment_definition: provided by user (optional)
  cohort_config: provided by user (optional)
  baseline_date: provided by user (optional)
Output: output/pm-metrics-ops/analysis-retention/
Validation: Retention calculated from full users not sampling; Cohort analysis covers time, channel, behavior dimensions; Aha Moment candidates pass significance testing; churn prediction model accuracy > 70%
Mode: AI

Invoke data-analysis-report

Invoke: ${data-analysis-report}
Input:
  funnel_analysis: output/pm-metrics-ops/analysis-funnel/
  retention_analysis: output/pm-metrics-ops/analysis-retention/
  anomaly_detection: output/pm-metrics-ops/analysis-anomaly/
  decision_dace: decision-dace -> decision_insight.json (optional)
  metrics_system: metrics-system -> metrics_system.json (optional)
  analysis_time_range: provided by user
  product_info: provided by user (optional)
Output: output/pm-metrics-ops/data-analysis-report/
Validation: Executive summary contains 3 key findings + Top 1 recommendation; core metrics dashboard complete; funnel analysis includes biggest drop-off point and improvement opportunities; retention analysis includes lifecycle stages; each insight has data facts + business implications; at least 3 action recommendations each with priority and validation method; data scope and limitations documented
Mode: AI->Human

Stage Summary (post_pipeline)

Follows the stage summary protocol in [orchestrator-protocol.md](../../../codex-templates/orchestrator-protocol.md).

| Parameter | Value | |------|-----| | Sub-Skill output path | output/pm-metrics-ops/ | | Summary output path | output/phase-reports/pm-metrics-ops/analysis-orchestrator.md |

Downstream connections: primary: decision-orchestrator (data analysis complete, convert analysis insights into actionable decisions) alternatives:

• target: experiment-orchestrator

reason: Analysis findings need A/B testing to validate hypotheses condition: Data analysis finds uncertain causal relationships needing experimental verification

• target: iteration-orchestrator

reason: Analysis conclusions directly impact iteration priorities condition: Data analysis produces clear iteration direction recommendations special_cases: []

Stage Gates

| Gate | Condition | Failure Handling | |------|------|------------| | Anomaly detection running 24/7 | Anomaly detection pipeline running continuously, no interruptions | Immediately fix detection pipeline, activate backup monitoring | | Funnel core path coverage | Core business funnel defined and data complete | Supplement funnel definition, ensure core path coverage | | Retention Aha Moment candidate identified | retention-analysis output file generated and non-empty | Expand behavior search scope or extend analysis period | | Data insight report generated | Data insight report file generated and non-empty | Supplement analysis or mark "recommend supplementing data" | | Stage summary generated | output/phase-reports/pm-metrics-ops/analysis-orchestrator.md generated and all 6 structural items non-empty | Supplement missing structural items and regenerate |

Human Decision Points

| Decision Point | Trigger Condition | Decision Content | |--------|----------|----------| | P0 anomaly immediate confirmation | P0-level anomaly detection triggered | Confirm anomaly authenticity, decide response strategy |

Decision Rules

| Condition | Action | |------|--------| | P0 anomaly | Immediate push + phone alert | | P1 anomaly | Slack/WeCom notification within 2 hours | | P2 anomaly | Daily summary report | | P3 fluctuation | Log only, no alert |

Exception Handling

| Exception Type | Handling Strategy | |----------|----------| | Sub-Skill output validation failed | Pause downstream stage execution, output validation failure report, prompt human to fix and retry current stage | | P0 anomaly detection triggered | Immediately interrupt current stage, prioritize P0 anomaly handling, resume original flow after handling | | Upstream data source unavailable | Execute per sub-Skill degradation strategy, record degradation info, mark degradation impact scope in final output | | Analysis results lack action recommendations | Block downstream transmission, require current sub-Skill to supplement action recommendations | | Human decision timeout without response | Pause flow, preserve current stage state, support resuming from checkpoint after human returns | | Stage summary generation failed | Generate partial summary based on completed sub-Skill outputs, mark missing items as "data missing", do not block orchestration completion |

Standalone Usage Input Acquisition Strategy

Standalone Trigger Scenario Identification

When this orchestrator is invoked directly (not through a parent orchestrator), it is considered a standalone trigger scenario. Typical trigger methods:

• User directly requests capabilities within this orchestrator's domain
• Triggered as an independent skill by external systems
• Parent orchestrator not executed, but user only needs this orchestrator's capability

Required Input Acquisition Strategy

| Required Input | Priority: Read from output/ | Fallback: Get from user conversation | Last Resort: AI knowledge base inference | |---------------|---------------------------|-------------------------------------|---------------------------------------| | PRD (prd.md) | Read output/pm-design/design-prd/prd.md | Ask user for PRD document or verbal requirements | Infer requirements from user description (low confidence, mark "PRD is AI-inferred") | | project_dir | — | Ask user for project directory path | Cannot infer, user must provide |

Upstream Orchestrator Auto-Backtracking

When critical required inputs are missing, suggest user execute upstream orchestrators in the following priority:

| Missing Input | Suggested Upstream Orchestrator | Description | |--------------|-------------------------------|-------------| | PRD | pm-design related orchestrator | PRD is the business basis for analysis-orchestrator, missing will result in execution without business foundation | | Sub-skill outputs | Sub-skill execution | Sub-skills (analysis-funnel, analysis-anomaly, analysis-retention...) produce domain-specific outputs |

Backtracking suggestion output format:

Critical input missing detected, suggest executing upstream orchestrator first:
1. [Priority] pm-design related orchestrator -> Produces PRD
Continue with AI-inferred values? (Inferred values confidence =0.5 | When confidence  project_dir validity -> Input confidence assessment

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [LuckyOneTwoThree](https://github.com/LuckyOneTwoThree)
- **Source:** [LuckyOneTwoThree/vibe-skill](https://github.com/LuckyOneTwoThree/vibe-skill)
- **License:** MIT
- **Homepage:** https://luckyonetwothree.github.io/all-skill-html/

Install and usage instructions live in the source repository linked above.