📦
planning-with-files
83Organize Long AI Tasks with Planning Files
Long AI sessions can lose goals, decisions, and progress. This skill guides Claude, Codex, and Claude Code to store plans, findings, and logs in project files.
Productivity View
Skills subagent-driven-development
📦
subagent-driven-development
Content revision r1 Safe ⚙️ External commands📁 Filesystem access
Coordinate Subagent Development Reviews
Independent implementation tasks can lose quality when context and review steps are inconsistent. This skill supplies a repeatable subagent workflow with implementer, spec review, and code quality checkpoints.
Supports: Claude Codex Code(CC)
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Agent request
Review the Skillstore skill "subagent-driven-development" from https://skillstore.io/skills/zhanlincui-subagent-driven-development.md and its manifest at https://skillstore.io/api/skills/zhanlincui-subagent-driven-development/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 "subagent-driven-development". A five-task feature plan with mostly independent tasks.
Expected outcome:
A task list, per-task implementer handoffs, spec review results, quality review results, and a final readiness check.
Using "subagent-driven-development". A task where the implementer added an unrequested option.
Expected outcome:
A spec review report naming the extra work, a fix request, a re-review result, and then a quality review handoff.
Using "subagent-driven-development". A task where quality review finds a maintainability issue.
Expected outcome:
A focused fix request to the implementer, a second quality review, and completion only after approval.
Security Audit
Safev5 • 7/8/2026 Open versioned report
The static external-command alerts are Markdown code fences or inline file references, not executable Ruby or shell commands. The hidden-directory and reconnaissance alerts are illustrative workflow text and review checklist language. No evidence of prompt injection, exfiltration, or malicious intent was found.
4
Files scanned
403
Lines analyzed
0
Review items
0
False positives ignored
Risk Factors
⚙️ External commands (8)
📁 Filesystem access (2)
No confirmed security findings were detected by the latest completed static and semantic audit. This does not prove the skill has no side effects.
Audited by: codex View Audit History →
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/zhanlincui-subagent-driven-development/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/zhanlincui-subagent-driven-development?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/zhanlincui-subagent-driven-development?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/zhanlincui-subagent-driven-development/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/zhanlincui-subagent-driven-development.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
ZhanlinCui. (2026). subagent-driven-development security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/zhanlincui-subagent-driven-development/audits/5BibTeX citation
@techreport{zhanlincui-zhanlincui-subagent-driven-development-2026,
author = {ZhanlinCui},
title = {subagent-driven-development security audit report (audit version 5)},
institution = {Skillstore},
year = {2026},
number = {5},
url = {https://skillstore.io/skills/zhanlincui-subagent-driven-development/audits/5},
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: "subagent-driven-development security audit report (audit version 5)"
version: "unspecified"
type: report
authors:
- name: "ZhanlinCui"
date-released: "2026-07-08"
url: "https://skillstore.io/skills/zhanlincui-subagent-driven-development/audits/5"
identifiers:
- type: other
value: "skillstore:zhanlincui-subagent-driven-development:audit:5"
description: "Skillstore immutable audit report identifier"
Compare variants
9 installable variantsEach author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.
Why this variant is first
Higher Skillstore usage
Author / source Skillstore Score Evidence Confidence Skillstore usage Updated
zhanlincui-subagent-driven-development
Skillstore Score 78
Evidence Confidence High
Skillstore usage 55
Updated
2026-08-21
obra-subagent-driven-development
Skillstore Score 78
Evidence Confidence High
Skillstore usage 15
Updated
2026-08-21
sickn33-subagent-driven-development
Skillstore Score 78
Evidence Confidence High
Skillstore usage 12
Updated
2026-08-21
cygnusfear-subagent-driven-development
Skillstore Score 78
Evidence Confidence High
Skillstore usage 7
Updated
2026-08-21
doyajin174-subagent-driven-development
Skillstore Score 78
Evidence Confidence High
Skillstore usage 6
Updated
2026-08-21
cycleaddict-subagent-driven-development
Skillstore Score 78
Evidence Confidence High
Skillstore usage 6
Updated
2026-08-21
dyai2025-subagent-driven-development
Skillstore Score 75
Evidence Confidence Medium
Skillstore usage 18
Updated
2026-08-21
codingcossack-subagent-driven-development
Skillstore Score 74
Evidence Confidence High
Skillstore usage 8
Updated
2026-08-21
davila7-subagent-driven-development
Skillstore Score 70
Evidence Confidence High
Skillstore usage 9
Updated
2026-08-21
Skillstore Score
Why this score Evidence Confidence: High 55
Architecture
85
Maintainability
87
Content
70
Community
83
Spec Compliance
What You Can Build
Execute planned feature work
Run a prepared feature plan through implementer and reviewer subagents without changing sessions.
Enforce specification compliance
Check that delivered code matches the task requirements before quality review begins.
Structure AI coding delegation
Give coding agents clear handoffs, review gates, and completion rules for multi-task changes.
Try These Prompts
Start a small plan
Use subagent-driven development for this plan. Extract each task, create a task list, and start with the first implementer subagent.
Run one task with review
Use the implementer prompt for Task 1. After completion, run the spec compliance reviewer, then the code quality reviewer if spec review passes.
Repair spec gaps
Send the spec reviewer findings back to the same implementer subagent. Re-run spec review after fixes before starting quality review.
Complete a multi-task branch
Continue the subagent-driven workflow for every remaining task. After all task reviews pass, run a final full implementation review.
Best Practices
- Paste the full task text into each implementer prompt instead of sending agents to read the whole plan.
- Run spec compliance review before code quality review for every task.
- Keep review loops open until the reviewer confirms the issue is fixed.
Avoid
- Dispatching multiple implementation subagents in parallel on the same codebase.
- Skipping spec review because the implementer report sounds complete.
- Starting the next task while review issues remain unresolved.
Frequently Asked Questions
What does this skill coordinate?
It coordinates implementer, spec reviewer, and code quality reviewer subagents for independent implementation tasks.
Does it write code by itself?
No. It provides workflow instructions and prompt templates for agents that perform the implementation and reviews.
When should I use it?
Use it when you have an implementation plan with tasks that can be completed one at a time.
When should I avoid it?
Avoid it when tasks are tightly coupled or when you need a planning discussion before implementation starts.
Why are there two review stages?
Spec review checks that the requested work was done. Quality review checks maintainability after the scope is correct.
Can this work with Claude, Codex, and Claude Code?
Yes. The workflow is written as prompt guidance for Claude, Codex, and Claude Code style agent delegation.
Developer Details
Author
ZhanlinCuiLicense
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Repository
https://github.com/ZhanlinCui/Ultimate-Agent-Skills-Collection/tree/main/subagent-driven-developmentRef
88a205c7f635a966e31156313b590d59007c5caa
Maintenance freshness
7/18/2026
Usage
46 downloads · 205 views
File structure
More from ZhanlinCui
View all📦
github-release-assistant
83Generate Bilingual GitHub Release READMEs
Release notes and README updates can take time before each GitHub release. This skill generates bilingual release documentation and guides review, commit, and push steps.
Documentation View
📦
obsidian-helper
81Organize Obsidian Notes with AI
Obsidian vaults can become hard to review and update consistently. This skill helps Claude and Codex create daily notes, capture ideas, and summarize recent work.
Productivity View
📦
Process PDFs with Extraction and Form Tools
PDF work often requires separate tools for text, tables, pages, and forms. This skill gives Claude, Codex, and Claude Code guided workflows and helper scripts for PDF tasks.
Office & Reports View
📦
frontend-design
79Build Distinctive Frontend Interfaces
Generic web interfaces can feel unfinished and forgettable. This skill guides Claude, Codex, and Claude Code toward polished frontend work with clear visual direction.
Design & Creative View
📦
brainstorming
79Plan Creative Work Before Building
Teams often start implementation before goals, constraints, and trade-offs are clear. This skill guides Claude, Codex, and Claude Code through structured brainstorming before design work begins.
Productivity View
Related Skills
View all📦
code-reviewer
83Review Code with Structured Quality Checks
by alirezarezvani
Code reviews can miss quality, security, and maintainability issues. This skill guides structured reviews with scripts, checklists, and report generation.
Coding & Development View
📦
lark-skill-maker
81Create Reusable Lark CLI Skills
by larksuite
Building reusable Lark automation requires consistent API research, permissions, and workflow documentation. This skill guides Claude and Codex through focused lark-cli skills for single APIs or multi-step workflows.
Coding & Development View
📦
cascade-orchestrator
81Orchestrate Multi-Model Skill Cascades
by DNYoussef
Complex AI work often needs several skills, models, and quality checks. This skill helps design cascades with routing, sandbox testing, swarm coordination, and memory.
Coding & Development View
📦
coding-agent-pm
81Manage Coding Agent Workflows
by kiwi-phantomworks
Coding-agent projects can drift when tasks, scope, and checks are vague. This skill gives PM templates and review rules that keep Claude, Codex, and Claude Code work bounded and verifiable.
Coding & Development View
📦
code-review
80Review Code for Bugs and Security Issues
by 21pounder
Code reviews can miss bugs, security flaws, and maintainability issues. This skill guides Claude, Codex, and Claude Code through structured review steps.
Coding & Development View
📦
vercel-react-best-practices
80Optimize React and Next.js Code
by vercel-labs
React and Next.js applications often lose performance through waterfalls, large bundles, and avoidable re-renders. This skill gives Claude, Codex, and Claude Code concrete review rules for faster frontend work.
Coding & Development View