Use when the user asks to "identify PM automation candidates", "automate PM reporting", "reduce manual PM processes", "find automation quick wins", or "design workflow automation". Activates when a stakeholder needs to scan PM processes for automation potential, calculate automation ROI, design automation specifications, prioritize automation backlog, or plan phased automation rollout across the PMO.
TL;DR: Systematically identifies PM process automation candidates across reporting, tracking, alerts, approvals, and data collection. Evaluates each candidate by ROI (hours saved vs. implementation effort), risk (automation failure impact), and feasibility (tool capability). Produces a prioritized automation backlog with specifications for the top candidates and a phased implementation roadmap.
Los PMs gastan el 40% de su tiempo en actividades que una máquina puede hacer mejor: consolidar reportes, calcular métricas, enviar recordatorios, actualizar dashboards. Cada hora automatizada es una hora devuelta a las actividades donde los humanos son insustituibles: negociación con stakeholders, resolución de conflictos, y toma de decisiones bajo incertidumbre.
# Scan all PM processes for automation opportunities
/pm:automation-opportunities $PROJECT --type=scan --scope="all"
# Calculate ROI for specific automation candidates
/pm:automation-opportunities $PROJECT --type=roi --candidates="reporting,metrics"
# Design automation specification for top candidates
/pm:automation-opportunities $PROJECT --type=spec --top=5Parameters:
| Parameter | Required | Description |
|---|---|---|
$PROJECT | Yes | Project or PMO identifier |
--type | Yes | scan, roi, spec, roadmap |
--scope | No | Process area filter (reporting, tracking, alerts, all) |
--candidates | No | Specific candidates for ROI analysis |
--top | No | Number of top candidates for specification |
{TIPO_PROYECTO} variants:
skills/automation-opportunities/references/*.md for automation pattern catalogsGood Automation Analysis:
| Attribute | Value |
|---|---|
| Processes inventoried | 25 PM activities across 6 categories |
| Candidates identified | 12, each with ROI calculation |
| Priority matrix | 2x2 with 4 quick wins identified |
| Top 5 specs | Trigger + Input + Logic + Output + Fallback |
| Pilot selected | 2 automations with success criteria |
| Roadmap | 3 phases over 6 months |
Bad Automation Analysis: A document that says "automate everything" without process inventory, ROI calculation, or feasibility check. No specifications for implementation. No pilot selection. Fails because it provides no actionable path from current manual state to automated state, and no evidence that automation will save more time than it costs to implement.
| Resource | When to read | Location |
|---|---|---|
| Body of Knowledge | Before starting to understand standards and frameworks | references/body-of-knowledge.md |
| State of the Art | When benchmarking against industry trends | references/state-of-the-art.md |
| Knowledge Graph | To understand skill dependencies and data flow | references/knowledge-graph.mmd |
| Use Case Prompts | For specific scenarios and prompt templates | prompts/use-case-prompts.md |
| Metaprompts | To enhance output quality and reduce bias | prompts/metaprompts.md |
| Sample Output | Reference for deliverable format and structure | examples/sample-output.md |
Prioritizes implementations. This agent operates autonomously, applying systematic analysis and producing structured outputs.
Scans for automatable tasks. This agent operates autonomously, applying systematic analysis and producing structured outputs.
Calculates automation ROI. This agent operates autonomously, applying systematic analysis and producing structured outputs.
Maps to available tools. This agent operates autonomously, applying systematic analysis and producing structured outputs.
15f6e4a
Also appears in
since Aug 28, 2026
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.