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.
npx skills add miticojo/ge-app-mcp --skill agent-workflow-architectgit clone --depth 1 https://github.com/miticojo/ge-app-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/miticojo/ge-app-mcp/agent-workflow-architect)<a href="https://agentmods.dev/skills/miticojo/ge-app-mcp/agent-workflow-architect"><img src="https://agentmods.dev/badge/skills/miticojo/ge-app-mcp/agent-workflow-architect/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/miticojo/ge-app-mcp/agent-workflow-architect"><img src="https://agentmods.dev/badge/skills/miticojo/ge-app-mcp/agent-workflow-architect.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00082 | $0.00852 |
| Opus 5.5 | $0.00033 | $0.00341 |
| Sonnet 5.5 | $0.00016 | $0.00170 |
| Haiku 4.5 | $0.00008 | $0.00085 |
Grade A, and why
agent-workflow-architect 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 13d 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.
How it starts
The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent & Skill Architect
1. Understand what exists
- Always inspect with
gemini_enterprise_inspect(target: "agents", withengine_idto list orengine_id+agent_idto get details): it isreadOnlyHint: true, so in Gemini Enterprise it runs in parallel with zero confirmation prompts. Reservegemini_enterprise_manage_agentsforcreate,update, anddelete(call mutating tools one at a time sequentially inside Gemini Enterprise). Agent types:lowCodeAgentDefinition(Agent Designer),a2aAgentDefinition,adkAgentDefinition,managedAgentDefinition(Google-made, read-only: never modify or recreate these). - For low-code agents, read the whole
nodesarray:root_agentroutes to the nodes listed insubAgentIds, and each node has its own instruction, model, tools and data stores.
2. Pick the agent type
| Need | agent_type |
Required inputs |
|---|---|---|
| Natural-language playbook, optionally with specialists | low-code |
instruction, optional model, datastores, tools, subagents |
| Existing service speaking the A2A protocol | a2a |
endpoint_url (https) |
| Agent built with ADK and deployed on Vertex AI Agent Engine | adk |
reasoning_engine (projects/.../locations/.../reasoningEngines/...) |
3. Multi-agent design (low-code)
- Keep the root agent a router: its instruction says when to delegate to each sub-agent.
- Give each sub-agent one responsibility, a clear
description, and only the data stores and tools it needs.
4. Governance
sharing_scope: start withPRIVATE, widen toRESTRICTEDorALL_USERSonly on request. Sharing with specific people and ownership transfer are done in the console; after an ownership transfer, scheduled and event triggers are disabled and must be re-enabled by the new owner.starter_prompts: 2-4 short examples of what the agent is for.- The lifecycle
state(ENABLED,PRIVATE,SUSPENDED, ...) is read-only: report it, do not try to set it.
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.
- 13d ago First seen · 54 lines · 82 tokens per session scan A 4a5c53d9282a
agent-workflow-architect is a skill published in the GitHub repository miticojo/ge-app-mcp (0 stars, last pushed 13d ago), licensed Apache-2.0. It adds 82 tokens to every session and 852 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-26.
agent-workflow-architect compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| agent-workflow-architectmiticojo/ge-app-mcp | Skill | 0 | 13d ago | Apache-2.0 | A | 82 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | yesterday | MIT | A | 200 | Claude Code, Codex |
| playwright-climicrosoft/playwright | Skill | 100k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 394k | today | MIT | A | 36 | Claude Code |
| systematic-debuggingobra/superpowers | Skill | 314k | 4d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |