skills.rest
hve4638
hve4638COMMUNITY
Explore all skills by this author →
GitHub

cross-review

Coordinate 1-to-1 cross-reviews between codex and subagents with sequential verification.

Updated Mar 1, 2026
Search Tags:#workflow#qa#verification#codex#review#subagent#cross-review
One-click install
npx skills add https://github.com/hve4638/hve-cc-marketplace --skill cross-review-hve4638
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cross-review
Source: https://github.com/hve4638/hve-cc-marketplace/tree/main/common/skills/cross-review
Command: npx skills add https://github.com/hve4638/hve-cc-marketplace --skill cross-review-hve4638

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Cross-review coordination between codex and subagents ensures mutual checks and reduces the risk of erroneous outputs by enabling structured, two-sided evaluation.

Core Features & Use Cases

  • 1-to-1 task pairing between codex and subagents to keep tasks aligned and clear.
  • Enforce identical numbers of codex and subagents to avoid imbalance in reviews.
  • Trigger a mutual verification step to converge on a correct, validated result.
  • Use Case: QA for generated code, data analysis, or content reviews that require cross-validation.

Quick Start

Propose a cross-review task with one codex task and one subagent task to validate the output.

Frequently Asked Questions about cross-review

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

FAQPage Schema
How do I set up mutual code reviews between codex and subagents?▼

Mutual code reviews between codex and subagents require proposing a 1-to-1 task pairing, maintaining equal numbers of codex and subagents via the codex MCP, and enforcing a sequential cross-verification flow.

What is cross-verification for parallel QA tasks?▼

Cross-verification for parallel QA tasks is a workflow that assigns paired checks to codex and subagents, triggering mutual verification steps to converge on a correct, validated output.

When should I use a 1-to-1 task pairing workflow for code QA?▼

Use a 1-to-1 task pairing workflow for code QA when you need structured, two-sided evaluation for generated code, data analysis, or content reviews that require cross-validation to reduce erroneous outputs.

Do I need the codex MCP to coordinate subagent reviews?▼

Yes, coordinating subagent reviews requires the codex MCP to maintain equal numbers of codex and subagents, ensuring tasks remain aligned and clear during the mutual verification process.

What are the limitations of enforcing sequential cross-verification flows?▼

Enforcing sequential cross-verification flows limits parallel execution speed, requiring strict 1-to-1 task pairing and identical numbers of codex and subagents to avoid imbalance and converge correctly.

Related Skills

Explore More Software Engineering Skills→
GenproMMGenproMM

gsd-execute-phase

Orchestrate phase execution across subagents using wave-based parallelism.

Official
Intermediate
mejangomejango

sarif-parsing

Parse SARIF 2.1.0 files to extract and deduplicate findings.

Community
Intermediate
eventua11yeventua11y
1

writing-tests

Automate E2E and unit test setup and execution with Playwright and Vitest.

Official
Intermediate
SighthesiaSighthesia
4

subagent-driven-development

Dispatch a fresh subagent per task and review code between tasks.

Community
Advanced