agentleFS
philpalmieriphilpalmieri/engineering-manager-skills/skills/metrics/review-cards/SKILL.md

review-cards

Skill1 starsChanged 6 months agoMITView on GitHub

Generate quarterly review cards grading each engineer across multiple dimensions: output, velocity, collaboration, quality, breadth, review depth, responsiveness, and mentorship. Grades are level-adjusted (expectations scale with seniority). Use when asked for review cards, quarterly grades, or individual performance cards.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If levels aren't in team.json, ask the user before proceeding

    line 141

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Engineer Review Cards

Generate grade-based review cards for each team member, assessing performance across multiple dimensions with level-adjusted expectations.

When to Trigger

  • "review cards", "quarterly grades", "performance cards"
  • "how is [person] doing?", "grade the team"
  • "level-adjusted performance", "expectations by level"
  • As part of a team health report

Data Sources

  • data/YYYY-MM/{login}-prs.json — output and velocity
  • data/YYYY-MM/{login}-reviews.json — collaboration and breadth
  • data/YYYY-MM/{login}-issues.json — issue engagement
  • data/enrichment/{pr_number}.json — quality (review rounds), review depth
  • team.json — level information for expectation scaling

Process

Step 1: Define level expectations

Expectations scale with level. Define thresholds per level:

DimensionJunior (L1-L2)Mid (L3-L4)Senior (L5-L6)Staff+ (L7+)
PRs/month5-128-2210-3010-30+
Reviews/month5-2015-5025-7025-70+
Repo breadth1-43-85-125-12+

These are configurable. If team.json has level data, map members to the appropriate tier. If levels are non-standard (e.g., "Senior", "IC3", "L9"), ask the user to confirm the mapping or infer from common patterns.

Step 2: Calculate raw metrics per person per quarter

MetricCalculation
OutputPRs authored per month (averaged over quarter)
VelocityMedian time-to-merge (hours)
CollaborationReviews given to teammates per month
QualityAverage review rounds before merge (lower = better)
BreadthDistinct repos with activity
Review Depth% of reviews with written comments or change requests (vs rubber stamps)
ResponsivenessMedian hours to first review on teammates' PRs
MentorshipReview count weighted 2× for reviews given to newer team members (< 12 months tenure)

Step 3: Assign grades

Map each metric to a grade based on level-adjusted expectations:

GradeMeaning
A+Significantly exceeding level expectations
AExceeding expectations
B+Meeting expectations with strength
BMeeting expectations
CSlightly below expectations
DBelow expectations

Grading uses the level-appropriate thresholds:

  • A+: > 130% of "exceeding" threshold
  • A: In "exceeding" range
  • B+: High end of "meeting" range
  • B: In "meeting" range
  • C: In "developing" range
  • D: Below "developing" range

Step 4: Calculate overall grade

Weighted average of dimensions:

  • Output: 20%
  • Velocity: 10%
  • Collaboration: 20%
  • Quality: 15%
  • Breadth: 10%
  • Review Depth: 10%
  • Responsiveness: 10%
  • Mentorship: 5%

Step 5: Growth trajectory

Compare current quarter to previous 1-2 quarters:

  • 📈 Growing: overall grade improved
  • ➡️ Stable: same grade (±0.3)
  • 📉 Declining: overall grade dropped

Step 6: Multi-quarter view

Show a table with grades across the last 3-4 quarters to visualize trends at a glance.

Output Format

# Engineer Review Cards: {quarter}

## Team Grading Summary
| Person | Level | Output | Velocity | Collab | Quality | Breadth | Depth | Response | Mentor | Overall | Trend |
|--------|-------|--------|----------|--------|---------|---------|-------|----------|--------|---------|-------|
| alice  | Sr    | A      | A+       | B      | A       | A       | B+    | A        | C      | **A-**  | 📈    |
| bob    | Mid   | B      | C        | A      | B       | B       | A     | B        | A      | **B+**  | ➡️    |

---

## Individual Cards

### alice (Senior Engineer)

| Quarter | Output | Velocity | Collab | Quality | Breadth | Depth | Response | Mentor | Overall |
|---------|--------|----------|--------|---------|---------|-------|----------|--------|---------|
| Q1      | B      | A        | C      | B       | B       | —     | —        | —      | B       |
| Q2      | A      | A        | B      | A       | A       | B     | A        | C      | A-      |
| Q3      | A      | A+       | B      | A       | A       | B+    | A        | C      | A-      |

**Growth trajectory:** 📈 Strong upward trend

**Strengths:** Velocity, output, breadth
**Growth areas:** Mentorship (low review investment in newer members)

**Domain expertise:**
- 🔵 Owner: auth-service, api-gateway
- 🟢 Contributor: shared-utils, deploy-infra

Notes

  • Quarters with < 1 month of data get "—" (insufficient)
  • New team members (< 3 months) get a ramp-adjusted grade (expectations halved)
  • The quality metric inverts: lower rounds = better grade
  • The velocity metric inverts: lower TTM = better grade
  • "—" for metrics where enrichment data isn't available
  • If levels aren't in team.json, ask the user before proceeding

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.