Team Analysis Skill
Overview
This skill helps surface hidden contributions and provides context for team member performance.
Core Principle: Hidden Impact
Traditional metrics miss the real story. Some people close 10 small tickets; others spend a week on complex architecture. Both are valuable.
Instructions
1. Never Judge by Ticket Counts Alone
When asked about someone's output, search for the fuller picture:
- •In-progress work (might be complex)
- •Comments and discussions
- •Related decisions
- •Blocked items
2. Search for Hidden Contributions
Use search_memories with query for the person's name plus terms like:
- •contribution, review, decision, help, unblock
Look for force-multiplier work: reviews, mentoring, unblocking others.
3. Provide Context, Not Judgement
Structure your response:
- •Visible work - What metrics show
- •Hidden impact - What metrics miss
- •Context - Why the picture looks this way
- •Assessment - Fair evaluation
4. Recognize Legitimate Patterns
- •Senior devs: more reviews, fewer individual tickets
- •Tech leads: architecture, unblocking, meetings
- •New hires: learning curve
- •Complex work: fewer items, longer duration
Example Response Format
code
[Name] this week: **Visible work:** - 2 tickets completed **Hidden impact:** - 4 code reviews unblocking others - Architecture discussion shaping API design - Helped debug auth issue (saved ~day) **Assessment:** Force-multiplier work that doesn't show in ticket counts.