agent-workflow

An automated development workflow that sends a feature idea through several specialized agents, from specification and architecture to coding, validation, and tests.

In plain words
What is it for?
It helps turn a requested feature into specifications, an architecture, implementation, quality checks, and a test suite through chained sub-agents.
Why use it?
It organizes multi-step development and repeats earlier work when validation finds that the quality threshold was not met.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/yonggao/claude-plugins/agent-workflow
Clone the repo
git clone --depth 1 https://github.com/yonggao/claude-plugins
Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,282 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00014 $0.01282
Opus 5 $0.00007 $0.00641
Sonnet 5 $0.00003 $0.00256
Haiku 4.5 $0.00001 $0.00128

Measured 2d ago against content hash c1f28642d514, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-workflow scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/agent-workflow.md · 177 lines

How it starts

The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Workflow - Automated Development Pipeline

Execute complete development workflow using intelligent sub-agent chaining with quality gates.

Context

  • Feature to develop: $ARGUMENTS
  • Automated multi-agent workflow with quality gates
  • Sub-agents work in independent contexts with smart chaining

Your Role

You are the Workflow Orchestrator managing an automated development pipeline using Claude Code Sub-Agents. You coordinate a quality-gated workflow that ensures 95%+ code quality through intelligent looping.

Sub-Agent Chain Process

Execute the following chain using Claude Code's sub-agent syntax:

First use the spec-analyst sub agent to generate complete specifications for [$ARGUMENTS], then use the spec-architect sub agent to design system architecture, then use the spec-developer sub agent to implement code based on specifications, then use the spec-validator sub agent to evaluate code quality with scoring, then if score ≥95% use the spec-tester sub agent to generate comprehensive test suite, otherwise first use the spec-analyst sub agent again to improve specifications based on validation feedback and repeat the chain.

Workflow Logic

Quality Gate Mechanism

  • Validation Score ≥95%: Proceed to spec-tester sub agent
  • Validation Score <95%: Loop back to spec-analyst sub agent with feedback
  • Maximum 3 iterations: Prevent infinite loops

Chain Execution Steps

  1. spec-analyst sub agent: Generate requirements.md, user-stories.md, acceptance-criteria.md
  2. spec-architect sub agent: Create architecture.md, api-spec.md, tech-stack.md
  3. spec-developer sub agent: Implement code based on specifications
  4. spec-validator sub agent: Multi-dimensional quality scoring (0-100%)
  5. Quality Gate Decision:
    • If ≥95%: Continue to spec-tester sub agent
    • If <95%: Return to spec-analyst sub agent with specific feedback
  6. spec-tester sub agent: Generate comprehensive test suite (final step)

Expected Iterations

Read the full file on GitHub · 177 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 177 lines · 14 tokens per session scan A c1f28642d514

Subscribe to this mod's changes

agent-workflow is a command published in the GitHub repository yonggao/claude-plugins (2 stars, last pushed 8mo ago), licensed MIT. It adds 14 tokens to every session and 1,282 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.