Code Review Standards Output Format

Code / PR Summary

One-sentence purpose and scope of changes or selected code.

Risk Assessment

Low / Medium / High, with a one-sentence justification.

Strengths

• What is excellent (bullet list)

Changed Files Overview

File	Lines Changed	Risk Level	Issues Found
path/to/file.ext	+12 -3	Medium	2

Assign risk levels based on component responsibility: High for files handling security, authentication, data persistence, or financial logic; Medium for core business logic and API boundaries; Low for utilities, configuration, and cosmetic changes.

Findings

Only include findings for lines present in the diff. Number findings sequentially and order by severity; Critical → High → Medium → Low.

Use the following format for each finding:

#### Issue {number}: [Brief descriptive title]

**Severity**: High / Medium / Low
**Category**: \<skill-defined category, e.g. Error Handling, Input Validation\>
**Skill**: \<exact skill `name` from frontmatter that surfaced this finding\>
**File**: `path/to/file.ext`
**Lines**: 45-52

### Problem

[Specific description of the issue and why it matters]

### Current Code

```language
[Exact code from the diff that has the issue]
```

### Suggested Fix

```language
[Exact replacement code that fixes the issue]
```

Findings Format Rules

• Group findings by severity: High -> Medium -> Low.
• When a skill defines its own findings format (e.g. a different table Layout), the agent's Output Format takes precedence.
• Omit the ### Current Code and ### Suggested Fix sections when no code change is needed (e.g. documentation-only findings).

Positive Changes

Highlight well-implemented patterns and improvements observed in the diff.

Testing Recommendations

List specific tests to add or update based on the review findings.

Recommended Actions

1. Must-fix before merge
2. Nice-to-have
3. Tooling / CI improvements

Acceptance Criteria Coverage (Story context only, included when story ID and definition are provided)

#	Acceptance Criterion	Status	Notes
1	...	Implemented / Partial / Not found	...

Out-of-scope Observations (pre-existing issues in unchanged code - excluded from verdict)

Issue	Recommendation
...	Consider fixing separately

Overall Verdict

Select based on the highest severity finding:

• Any Critical or High findings → ❌ Request changes
• Only Medium or Low findings → 💬 Approve with comments
• No findings → ✅ Approve

---

Skills Loaded: <comma-separated list of loaded skill names> Skills Unavailable: <comma-separated list of skill names whose SKILL.md was not found at the expected path, or "none">

---

🤖 Crafted with precision by ✨Copilot following brilliant human instruction, then carefully refined by our team of discerning human reviewers.