4-step-program
Orchestrate Code Review Loops
Coordinating AI coding agents can leave gaps in issue coverage and reviews. This skill gives Claude, Codex, and Claude Code a strict loop for delegation, review, repair, and PR presentation.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "4-step-program" from https://skillstore.io/skills/cygnusfear-4-step-program.md and its manifest at https://skillstore.io/api/skills/cygnusfear-4-step-program/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "4-step-program". Coordinate an AI agent fixing an authentication bug.
Expected outcome:
A delegation summary with every issue requirement, a review gate, a final coverage assessment, and a clickable PR link.
Using "4-step-program". Check whether an agent can present a feature as complete.
Expected outcome:
- Coverage is confirmed for each requirement.
- The review is posted to the PR.
- No suggestions remain before presentation.
Using "4-step-program". Prepare final status after a reviewed PR.
Expected outcome:
A concise human report with completed work, requirement coverage, PR URL, related issue URLs, and remaining risks.
Security Audit
Medium RiskStatic findings are false positives from Markdown code fences, inline examples, placeholder GitHub URLs, and Mermaid diagram text. No prompt injection attempt was found, but the skill intentionally requires GitHub PR review posting, which creates external write side effects that need user authorization.
Confirmed security concerns (1)
Risk Factors
⚙️ External commands (57)
🌐 Network access (7)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/cygnusfear-4-step-program/audits/8?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/cygnusfear-4-step-program?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/cygnusfear-4-step-program?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/cygnusfear-4-step-program/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/cygnusfear-4-step-program.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
Cygnusfear. (2026). 4-step-program security audit report (audit version 8) [Author version unspecified]. Skillstore. https://skillstore.io/skills/cygnusfear-4-step-program/audits/8BibTeX citation
@techreport{cygnusfear-cygnusfear-4-step-program-2026,
author = {Cygnusfear},
title = {4-step-program security audit report (audit version 8)},
institution = {Skillstore},
year = {2026},
number = {8},
url = {https://skillstore.io/skills/cygnusfear-4-step-program/audits/8},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "4-step-program security audit report (audit version 8)"
version: "unspecified"
type: report
authors:
- name: "Cygnusfear"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/cygnusfear-4-step-program/audits/8"
identifiers:
- type: other
value: "skillstore:cygnusfear-4-step-program:audit:8"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: HighWhat You Can Build
Coordinate Bug Fixes
Manage AI-assisted bug fixes while preserving complete issue coverage and visible PR review history.
Standardize Agent Reviews
Require every code change to pass review, correction, and final evidence checks before presentation.
Build Repeatable Workflows
Create a consistent delegation pattern for Claude, Codex, or Claude Code coding agents.
Try These Prompts
Use the 4-step-program to coordinate a bug fix for Issue #[number]. Include all issue requirements, require a PR, and report the PR URL.
Apply the 4-step-program to a feature request. Extract every requirement, delegate implementation, require code-reviewer review, and loop until there are zero suggestions.
Coordinate two related issues with the 4-step-program. Link each issue in the PR body, verify closing references, and present coverage evidence.
Run the full 4-step-program for a complex refactor. Use a final requirement table, require GitHub review posting, include diagrams when useful, and escalate after five failed iterations.
Best Practices
- Use the skill only when the task should produce reviewed code changes.
- Verify issue requirements directly before delegating and before presenting results.
- Confirm that GitHub posting is permitted before asking agents to comment on PRs.
Avoid
- Using the workflow for research-only tasks that do not need a PR.
- Accepting a review with suggestions because tests passed.
- Posting repository details externally without confirming user approval.
Frequently Asked Questions
Does this skill write code?
Does it require GitHub?
Can it be used with Claude, Codex, and Claude Code?
What makes it different from a normal review prompt?
What should I check before using it?
What happens if the review is not perfect?
Developer Details
Author
CygnusfearLicense
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Ref
7e9e368dc1a370f2040369ebcd0db06d90cb48e5
Maintenance freshness
7/18/2026
Usage
5 downloads · 247 views
File structure
📄 SKILL.md