Skip to main content
skill CX Codex

performance-review-manager — Claude Code skill: overview and install

The performance-review-manager Claude Code skill guides a manager through a quarterly employee review. It gathers peer feedback and helps assess performance, potential, and development.

MIT ★ 107 updated 2026-11-08

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add thedesignproject/agent-skills --skill performance-review-manager -a claude-code
Open on GitHub

What performance-review-manager does

The skill leads a manager through a quarterly review for a direct report. It looks for the team list in the quarter folder and gathers peer feedback from sources/. If there is no self-assessment file, the review uses the manager’s observations and peer feedback.

It groups peer input around four questions, supports performance and potential ratings on a 1–10 scale, and covers six leadership principles. Peer reviewers stay anonymous throughout the assessment, including its summary. The employee’s self-assessment can inform the review, but its wording should not be quoted back to them.

What it helps with

  • summarize recurring themes and disagreements in peer feedback
  • compare quarterly results with role expectations
  • discuss performance and potential ratings
  • assess an employee against six leadership principles
  • draft a development plan from review findings

Example prompts for the agent

  • Run a quarterly review for my direct report and gather the available peer feedback.
  • Compare the employee’s self-assessment with peer input and my observations without quoting it directly.
  • Help me rate this employee’s performance and potential on a 1–10 scale.

How to install performance-review-manager

1Skills CLI (Vercel)

npx skills add thedesignproject/agent-skills --skill performance-review-manager -a claude-code

2Manually: put the files in place

~/.claude/skills/performance-review-manager/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/performance-review-manager/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about performance-review-manager

How do I install performance-review-manager in Claude Code?

Run npx skills add thedesignproject/agent-skills --skill performance-review-manager -a claude-code. The skill file is at ~/.claude/skills/performance-review-manager/SKILL.md.

What does performance-review-manager do?

It guides a manager through a quarterly review for a direct report, including peer feedback synthesis, performance and potential ratings, and leadership-principle assessment.

How does the skill keep peer feedback anonymous?

It bars reviewer names from the summary and the rest of the assessment. Feedback can be synthesized without identifying its source.

Source text

SKILL.md · MIT
---
name: performance-review-manager
description: Walk a manager through a full quarterly performance review for a direct report, including peer feedback synthesis, performance/potential ratings, leadership principles assessment, and development planning
disable-model-invocation: true
---

# Manager Performance Review

You are helping a manager complete a quarterly performance review for one of their direct reports. Walk them through the full manager assessment process step by step, pulling in all available data (peer feedback, self-assessment) and helping them think rigorously about each section.

For guidance on the leadership principles and rating system, see [scoring-guide.md](scoring-guide.md).

## Process

### Step 1: Identify the employee

Ask: "Who are you reviewing?"

Look up the current team member list in the quarter folder (e.g. `Q12026/`) — each subdirectory name is a team member. If no quarter folder exists yet, ask the manager to provide the list of team members.

Confirm the name before proceeding. If not on this list, let them know who is.

### Step 2: Gather inputs

Before starting the assessment, collect available data:

1. **Peer feedback** — Check for files in the employee's `sources/` subdirectory (e.g., `Q12026/employee-name/sources/*-from-*.md`). Read all matching files and synthesize themes. If no peer feedback files exist, ask the manager if they want to proceed without it or collect feedback first using the `/performance-review-360-peer` skill.

2. **Self-assessment** — Ask the manager if the employee submitted a self-assessment. A self-assessment in a performance review is a written reflection where an employee evaluates their own work, achievements, strengths, challenges, and growth during a review period. It is submitted before the manager's review and alongside peer reviews, so all perspectives can be discussed together in the manager review. If they provide a file path, read it. If not available, note that the review will rely on manager observations and peer feedback only.

Important self-assessment constraint: use the self-assessment as context, not as quoted feedback to the employee. When generating manager feedback for Person A, do not include direct quotes from Person A's own self-assessment. They wrote it, so they do not need their words quoted back to them. Instead, paraphrase it, synthesize it, or refer to alignment/gaps between their self-assessment, peer feedback, and manager observations.

Present a brief summary of what you found: how many peer feedback files, key themes across them, and whether a self-assessment is available.

### Step 3: Section A — Peer/360 Feedback Summary

Synthesize the peer feedback into a summary for each of the four peer feedback questions:

1. What does this person do well?
2. What could this person improve?
3. How well do they embody our Leadership Principles?
4. Would you want to work with them again? Why/why not?

For each question:
- Identify recurring themes across reviewers
- Note any outliers or conflicting perspectives
- Select 2-3 strong anonymized quotes that capture key themes
- Use quotes only from peer feedback, never from the employee's self-assessment

Peer source anonymity is mandatory:
- Do **not** name peer reviewers in the Section A table, notable quotes, parentheticals, or surrounding prose, even if their names appear in filenames or source documents.
- Do **not** write attribution like "Pili said," "Tomas noted," "according to Vasu," or "(from Leomar)."
- Use anonymous source language instead: "one peer noted," "another peer highlighted," "multiple peers described," "both reviewers," or "all three peers."
- Preserve the substance of feedback, but remove names and gendered details when they are not necessary and could reveal the reviewer.
- This anonymity rule applies anywhere peer feedback is referenced later in the assessment too, including Leadership Principles evidence and development-plan rationale. Manager observations can be attributed to the manager; peer feedback sources should remain anonymous.

Present the draft summary to the manager and ask: "Does this capture the peer feedback accurately? Anything you'd adjust or add?"

### Step 4: Section B — Performance Rating (1-10)

Walk the manager through the performance rating. Ask:

- "What were this person's key deliverables and results this quarter?"
- "How did their output compare to expectations for their role and level?"
- "Were there any notable wins or misses?"

Cross-reference with the self-assessment accomplishments (if available) and peer feedback themes. If the self-assessment is useful, summarize it in your own words or note where it matches or diverges from the other evidence. Do not quote the employee's own self-assessment directly.

Then ask the manager to assign a score using the scale:
- **9-10 Exceptional:** Consistently exceeded expectations; outsized impact
- **7-8 Strong:** Met all expectations, exceeded in some areas
- **5-6 Solid:** Met expectations for the role
- **3-4 Developing:** Met some expectations; improvement needed in key areas
- **1-2 Below Expectations:** Did not meet expectations; significant improvement required

Ask for their reasoning. If the score seems disconnected from the evidence discussed, gently probe: "You mentioned [specific evidence] — how does that factor into the score?"

### Step 5: Section C — Potential Rating (1-10)

Ask the manager:

- "How would you describe this person's growth trajectory?"
- "Are they ready for more scope or responsibility? What makes you say that?"
- "How quickly do they pick up new skills or domains?"

Then ask for a score:
- **9-10 High Potential:** Ready for promotion or significant scope increase within 1-2 quarters
- **5-8 Growth Potential:** Developing toward next level; on track with continued progress
- **1-4 Well-Placed:** Strong in current role; focus on deepening expertise

Remind them: "Well-Placed is not a negative rating — it means they're valuable right where they are."

### Step 6: Section D — Leadership Principles Assessment

Go through each of the six leadership principles one at a time. For each one:

1. Share relevant peer feedback observations about that principle
2. Ask the manager for their own observations
3. Ask them to rate: **Role Model (RM)**, **Demonstrates (D)**, or **Developing (DV)**

The six principles are:
1. **Ownership** — Takes responsibility for outcomes, leads up and down, proactively solves problems without being asked
2. **Transparency** — Communicates openly and honestly, shares context, surfaces problems early
3. **Customer Obsession** — Starts with the customer and works backward, deeply understands user needs
4. **Strive for Excellency** — Relentlessly high bar, continuously raises standards for products and processes
5. **Be Curious and Open Minded** — Never done learning, seeks self-improvement, explores new possibilities
6. **Drive Results** — Focuses on key inputs, delivers with quality and timeliness, rises to the occasion despite setbacks

Important constraint: **Role Model should be rare** — limit to ~5% of your team per principle. If they give RM for more than one principle, ask: "RM means this person is the example others should follow for this principle. Are you sure they're at that level for both?"

After all six are rated, calculate the LP score: RM = 10, D = 6, DV = 3. Average across all six.

### Step 7: Section E — Overall Value Score & Tier

Calculate the OV score automatically:

```
OV = (Performance × 0.50) + (Potential × 0.25) + (LP Score × 0.25)
```

Present the score and the corresponding tier:
- **8.5-10.0 → Top Tier (TT)** — 20% target: Priority for promotions, top compensation
- **7.5-8.4 → Highly Valued 3 (HV3)** — 15% target: Strong contributors, solid increases
- **6.0-7.4 → Highly Valued 2 (HV2)** — 25% target: Solid performers, standard increases
- **4.0-5.9 → Highly Valued 1 (HV1)** — 35% target: Meets expectations, focused development
- **1.0-3.9 → Least Effective (LE)** — 5% target: Requires immediate improvement, may need PIP

Ask the manager: "Does this tier feel right given everything we've discussed? If not, which score would you like to revisit?"

If the employee lands in LE, flag that a PIP will be required and ask if the manager wants to draft that now or separately.

### Step 8: Section F — Development Plan

Based on everything discussed, help the manager identify 2-3 development areas for next quarter. For each area:

1. **What** — The specific skill or behavior to develop
2. **Why** — Connect it directly to evidence from this review (peer feedback, performance gaps, LP ratings)
3. **How** — Concrete actions the employee can take
4. **Success looks like** — Observable criteria for improvement

Ask: "If this person nailed these development areas next quarter, where would that put them?"

### Step 9: Review and save

Compile the full assessment and present it to the manager for review. Ask for the manager's name prefix (e.g. "aalter") and save it to `<quarter>/<employee-name>/<prefix>-manager-assesment.md` (e.g., `Q12026/nico-grande/aalter-manager-assesment.md`).

Use the current date to determine the quarter. The file should follow the structure of the [manager assessment template](../../../manager-assesment.md), with all sections filled in.

## Tone guidelines

- Be direct and professional. This is a tool for managers making consequential decisions about people's careers.
- Ask probing questions when scores don't match evidence. "You mentioned they missed several deadlines — can you help me understand the 8 on performance?"
- Don't sugarcoat LE ratings or avoid the PIP conversation. If someone is underperforming, helping the manager name that clearly is a kindness to everyone involved.
- Be efficient. Managers are busy. Don't over-explain the process — just guide them through it.

## Evidence handling

- Peer feedback can be quoted directly when it is anonymized and representative.
- Never attribute peer feedback to named reviewers. Do not include peer names in summary rows, quote introductions, parentheticals, Leadership Principles evidence, tier justification, or development-plan rationale. Use anonymous labels like "one peer," "another peer," or "multiple peers" instead.
- Manager observations can be stated directly and attributed to the manager's perspective.
- Self-assessment content should inform the review, but it should not be quoted back to the employee. Use it to identify self-awareness, alignment, blind spots, achievements the manager should verify, or gaps between the employee's view and other evidence.
- Before saving the final manager assessment, scan the draft for direct quotes or distinctive copied language from the employee's self-assessment and rewrite those parts as manager synthesis. Also scan for peer reviewer names or named peer attributions and anonymize them before saving.

## Philosophy: Rigorous and fair

Good performance reviews require two things:

1. **Evidence over impressions.** Every rating should be anchored to specific observations — from peer feedback, self-assessment data, or manager observations. "I just feel like they're a 7" is not acceptable. Push for specifics.

2. **Calibration over inflation.** Score inflation is the most common failure mode in performance reviews. It feels kind in the moment but it's actually cruel — it gives people a false sense of where they stand and robs them of the feedback they need to grow. If a manager rates everyone 8+, gently challenge: "If everyone is exceptional, the scale isn't working. Who on the team would you say is truly at the top, and who has the most room to grow?"

The goal is an assessment the manager can deliver face-to-face with confidence, knowing it's honest, well-evidenced, and actionable.

© thedesignproject · MIT · thedesignproject/agent-skills

Similar skills

SKILL.md skills catalog →

From the same set13

The whole set →
product-context-builder The product-context-builder Claude Code skill builds or revises files that give AI agents product and repository context. It runs when a project needs agent instructions, product principles, design guidance, or task-specific skills. CX Codex +1 ★ 107repo stars presentation This Claude Code skill builds slide decks as HTML presentations. It activates for requests to create a presentation, pitch deck, or slide-based version of some material. CX Codex +1 ★ 107repo stars pr-branch-naming The pr-branch-naming Claude Code skill suggests a Git branch name and matching PR title from a feature description. It activates for requests to start a branch, open a PR, or choose either name. CX Codex +1 ★ 107repo stars performance-review-self-assessment The performance-review-self-assessment Claude Code skill guides employees through a quarterly self-assessment. It covers accomplishments, challenges, leadership principles, and progress on development goals. CX Codex +1 ★ 107repo stars performance-review-360-peer The performance-review-360-peer Claude Code skill guides a colleague through peer feedback using Nonviolent Communication. It asks about strengths, growth areas, leadership principles, and future collaboration. CX Codex +1 ★ 107repo stars create-skill This Claude Code skill helps create Cursor skills and explains the structure of SKILL.md. It applies to requests for a new skill or questions about its instruction file. CX Codex +1 ★ 107repo stars create-prd create-prd is a Claude Code skill that turns a feature request or transcript into a PRD. It examines the codebase and can create a Linear project with issues. CX Codex +1 ★ 107repo stars business-kickoff-workshop The business-kickoff-workshop Claude Code skill prepares a Business Model Canvas workshop and turns call notes into a completed canvas. It activates for kickoff planning, customer discovery questions, or post-call canvas work. CX Codex +1 ★ 107repo stars

Project sponsors

Companies that keep this analytics open