Installs into .claude/skills of the current project.
Are you the author of Portfolio?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/sebs-portfolio)
---
name: portfolio
description: Cross-project portfolio view comparing value distribution and effort allocation
allowed-tools: mcp__rewelo__project_list, mcp__rewelo__ticket_list, mcp__rewelo__calc_priority, mcp__rewelo__report_summary
---
# Multi-Project Portfolio View
Generate a cross-project portfolio analysis.
## Data collection
1. Use `project_list` to get all projects
2. For each project, get the **active tickets** with `ticket_list` and `excludeTags: ["state:done"]`: its items carry value, cost and priority (`ticket_list` returns 100 tickets per call: page with `offset` until you have `total` of them). `calc_priority` ranks done tickets too, so don't take the metrics below from it
3. Use `report_summary` for each project
## Portfolio metrics
For each project, calculate:
- **Total Value**: sum of (benefit + penalty) across active tickets
- **Total Cost**: sum of (estimate + risk) across active tickets
- **Average Priority**: mean priority of the active tickets
- **Ticket Count**: total active tickets
- **Completion Rate**: tickets tagged `state:done` / total tickets
## Output
### Project comparison table
| Project | Tickets | Total Value | Total Cost | Avg Priority | Done % |
|---------|---------|-------------|------------|-------------|--------|
### Value vs. Cost chart (text-based)
Show a simple bar comparison for each project:
```
ProjectA: Value ████████████ 120 Cost ██████ 60
ProjectB: Value ████ 40 Cost ████████ 80
```
### Imbalance analysis
Identify projects where:
- High value but low effort allocated (under-invested)
- Low value but high effort allocated (over-invested)
- High average priority (urgent backlog)
- Low completion rate (execution problems)
### Recommendations
Suggest rebalancing actions:
- "Project Alpha has 60% of total value but only 30% of effort. Consider shifting resources."
- "Project Beta has low value and high cost. Evaluate whether to continue or descope."