hoavdc/CodexKit / skills/codexkit-performance-review-writer
codexkit-performance-review-writer
VerifiedWrite balanced, evidence-based performance reviews for employees. Covers achievements, development areas, and goal setting using the SBI (Situation-Behavior-Impact) feedback model. Produces reviews that are fair, legally defensible, and growth-oriented. Use during annual or semi-annual review cycles. Do not use for disciplinary actions, PIPs, or compensation calculations.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.2 KB |
| agents/openai.yaml | config | 178 B |
| examples/common-mistakes.md | markdown | 687 B |
| examples/good-output.md | markdown | 1.7 KB |
| verification/checklist.md | markdown | 1.4 KB |
SKILL.md
View on GitHub---
name: codexkit-performance-review-writer
description: Write balanced, evidence-based performance reviews for employees. Covers achievements, development areas, and goal setting using the SBI (Situation-Behavior-Impact) feedback model. Produces reviews that are fair, legally defensible, and growth-oriented. Use during annual or semi-annual review cycles. Do not use for disciplinary actions, PIPs, or compensation calculations.
version: 1.0.0
category: review
---
# Performance Review Writer
## Purpose
Produce structured, balanced performance reviews that celebrate achievements, identify growth areas, and set actionable development goals — grounded in specific evidence, not subjective impressions.
## When to use
- annual or semi-annual performance review cycle
- probation-end evaluation for new hires
- mid-year check-in that requires written documentation
- when managers need help articulating balanced feedback
## When not to use
- Performance Improvement Plans (PIP) — those have a different legal format
- disciplinary warnings or termination documentation
- compensation or bonus calculations
## Inputs
- employee name, role, and review period
- key achievements and contributions during the period
- areas where performance fell short of expectations
- specific examples or evidence for both strengths and gaps
- goals from the previous review period and their status
- team context (team size, challenges, market conditions)
- desired development direction (promotion track, lateral move, skill building)
## Procedure
1. **Summarize overall performance** in 2–3 sentences:
- State the overall assessment clearly (Exceeds / Meets / Below)
- Ground it in outcomes, not personality
2. **Document achievements** using SBI model (3–5 items):
- Situation: the context or challenge
- Behavior: what the person specifically did
- Impact: the measurable result or business outcome
3. **Document development areas** using SBI model (2–3 items):
- Same structure but framed constructively
- Focus on behaviors that can be changed, not personality traits
- Avoid hedging language — be specific and honest
4. **Review previous goals**:
- Status each goal: Achieved / Partially Achieved / Not Achieved
- Note context that affected goal achievement
5. **Set new goals** for the next period (3–5 goals):
- SMART format: Specific, Measurable, Achievable, Relevant, Time-bound
- Mix of performance goals and development goals
- At least one stretch goal
6. **Write development recommendations**:
- Training, mentoring, stretch assignments, or lateral exposure
- Connect to career aspirations
7. **Bias check** — review the draft for:
- Recency bias (only citing recent events)
- Halo/horn effect (one trait coloring everything)
- Gendered language patterns
- Vague adjectives without evidence
## Output
- overall performance summary with rating
- achievements section (SBI format, 3–5 items)
- development areas section (SBI format, 2–3 items)
- previous goal status review
- new goals for next period (SMART format)
- development recommendations
- bias check notes
## Definition of done
- every achievement and development area uses SBI format with specific evidence
- overall rating is supported by the documented evidence
- new goals are in SMART format
- bias check is completed and documented
- tone is professional, growth-oriented, and legally defensible
## Examples
- "Write an annual review for a mid-level developer who exceeded delivery targets but needs to improve documentation."
- "Draft a probation review for a new marketing coordinator — strong start but inconsistent follow-through."
- "Help me make this review more specific — I wrote 'good team player' but need evidence."
## Quality Criteria
- [ ] Every strength and development area is supported by specific evidence.
- [ ] Feedback focuses on observable work behaviors and outcomes, not personality labels.
- [ ] The review balances recognition, accountability, growth, and future expectations.
- [ ] Goals are SMART and linked to role expectations or career path.
- [ ] Bias check covers recency, halo/horn effect, vague adjectives, and inconsistent standards.
## Verification (4C)
| Check | Question |
|-------|----------|
| **Correctness** | Are achievements, gaps, ratings, and goals supported by the evidence provided? |
| **Completeness** | Are achievements, development areas, prior goals, new goals, and bias checks included? |
| **Context-fit** | Does the review match the role level, review period, and performance expectations? |
| **Consequence** | Could the wording create unfairness, legal risk, or unclear expectations for the employee? |
## Edge Cases
- **Insufficient evidence** — Ask for examples before writing conclusions; do not fill gaps with assumptions.
- **Potential disciplinary issue** — Do not draft a PIP or warning letter; recommend HR review.
- **Promotion or compensation implications** — Keep review language evidence-based and avoid compensation recommendations.
- **Sensitive protected-class context** — Remove irrelevant personal details and require HR/legal review when needed.
## Changelog
- v1.1.0 — Added performance-review-specific quality gates, 4C checks, and HR safety edge cases.
- v1.0.0 — Initial release
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |