deepwork-receiving-code-review

Evaluates code review feedback against evidence before implementing corrections.

Updated Jun 18, 2026
One-click install
npx skills add https://github.com/hugefiver/ocmm --skill deepwork-receiving-code-review-hugefiver
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deepwork-receiving-code-review
Source: https://github.com/hugefiver/ocmm/tree/main/plugins/deepwork/skills/deepwork-receiving-code-review
Command: npx skills add https://github.com/hugefiver/ocmm --skill deepwork-receiving-code-review-hugefiver

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Code review feedback often arrives unclear, incorrect, or scope-expanding, and blindly implementing it introduces regressions while performative agreement erodes technical rigor. This Skill enforces verification-first handling of review findings so only justified corrections are applied. ## Core Features & Use Cases - Evidence-Based Evaluation: Classifies each finding as a product defect, evidence gap, optional improvement, incorrect finding, or intent-changing proposal before acting. - Source-Specific Handling: Applies different rules for user direction versus external reviewer suggestions, with pushback supported by code, tests, or platform evidence. - Scoped Follow-Up Review: Reruns only review lanes affected by substantive changes, avoiding redundant full re-reviews. - Use Case: A reviewer claims a function breaks API compatibility; the agent verifies the claim against current code and requirements, then either applies the minimal fix or pushes back with concrete evidence. ## Quick Start Use the deepwork-receiving-code-review skill to evaluate this review feedback and implement only the verified findings.

Frequently Asked Questions about deepwork-receiving-code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I handle code review feedback that seems incorrect?▼

Verify the claim against current code, requirements, and test evidence before implementing anything. If the finding is wrong, push back with concrete evidence such as code references, test results, or platform documentation rather than agreeing performatively.

How to decide which code review comments to implement?▼

Classify each finding as a product defect, evidence gap, optional improvement, incorrect finding, or intent-changing proposal. Implement defects with the smallest justified correction, gather evidence for gaps, defer optional items, and escalate proposals that change approved scope.

When should review feedback be escalated instead of implemented?▼

Escalate when a suggestion changes scope or acceptance criteria, weakens security or data guarantees, modifies a public API, expands permissions, requires irreversible actions, or conflicts with approved user direction. Material ambiguity without available evidence also warrants escalation rather than guessing.

Does every code review finding require rerunning the full review?▼

No. Rerun only the review lanes affected by substantive changes to code, requirements, acceptance criteria, or evidence. Formatting, narration, label changes, and unrelated edits do not require repeated review cycles.

What are the limitations of automated code review handling?▼

Reviewer approval labels and acknowledgements are informational only and do not substitute for verification. The approach cannot resolve material ambiguity without evidence, and it never expands Git or permission boundaries beyond explicitly authorized operations.