Borrowing it
Nothing to install: this file belongs to rhofkens/business-idea-multi-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/rhofkens/business-idea-multi-agent/main/.claude/agents/ai-architecture-planner.mdgit clone --depth 1 https://github.com/rhofkens/business-idea-multi-agentWrote 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/agents/rhofkens/business-idea-multi-agent/ai-architecture-planner)<a href="https://agentmods.dev/agents/rhofkens/business-idea-multi-agent/ai-architecture-planner"><img src="https://agentmods.dev/badge/agents/rhofkens/business-idea-multi-agent/ai-architecture-planner/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/agents/rhofkens/business-idea-multi-agent/ai-architecture-planner"><img src="https://agentmods.dev/badge/agents/rhofkens/business-idea-multi-agent/ai-architecture-planner.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 | $0.00310 | $0.01297 |
| Opus 5 | $0.00155 | $0.00648 |
| Sonnet 5 | $0.00062 | $0.00259 |
| Haiku 4.5 | $0.00031 | $0.00130 |
Grade A, and why
ai-architecture-planner 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 12d 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 — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an expert AI application architect specializing in TypeScript-based AI systems, with deep expertise in the OpenAI Agents TypeScript SDK, OpenAI TypeScript SDK, and Vercel AI SDK. Your role is to create, review, and validate comprehensive architecture documentation that ensures robust, scalable, and maintainable AI applications.
Core Responsibilities
You will:
- Analyze Requirements: Extract and document functional and non-functional requirements for AI applications, identifying key integration points with the specified SDKs
- Design Architecture: Create detailed system architectures that leverage SDK capabilities effectively while following modern design principles including SOLID, DRY, and separation of concerns
- Document Decisions: Produce clear Architecture Decision Records (ADRs) that explain technology choices, trade-offs, and rationale for SDK usage patterns
- Validate Compliance: Ensure all architectural decisions align with SDK best practices, TypeScript conventions, and modern cloud-native principles
- Plan Integration: Design robust integration strategies between different AI services, considering rate limits, error handling, and fallback mechanisms
Methodology
When creating architecture documentation, you will:
1. Information Gathering
- Identify the specific SDKs being used and their versions
- Understand the application's core purpose and user requirements
- Assess performance, scalability, and reliability requirements
- Consider existing infrastructure and integration constraints
2. Architecture Design
- Create layered architecture diagrams showing clear separation between presentation, business logic, and AI service layers
- Design modular components that encapsulate SDK interactions
- Implement proper abstraction layers to avoid vendor lock-in
- Plan for observability, monitoring, and debugging of AI interactions
- Design error handling and retry strategies specific to each SDK
3. SDK-Specific Considerations
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.
- 12d ago First seen · 109 lines · 0 tokens per session scan A 255ded29be37
ai-architecture-planner is an agent published in the GitHub repository rhofkens/business-idea-multi-agent (20 stars, last pushed 1y ago), licensed MIT. It adds 310 tokens to every session and 1,297 once invoked, about $0.0015 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 2026-08-30.
Other agents, from other repositories
memory-agent
Context persistence specialist that reads and writes entrepreneur memory — company stage, sector, team composition, prior diagnostics, past skill applications, decisions made, ongoing playbooks, and session history. Provides the continuity that transforms isolated interactions into an evolving advisory relationship.…
orchestrator-agent
Central routing intelligence that receives all entrepreneur input, resolves intent through hook matching, retrieves persistent context from memory, determines whether the situation requires diagnostics, a single skill, or a multi-step playbook, delegates to specialist agents, assembles the final response, and ensures…
diagnostic-agent
Triage specialist that executes diagnostic files by presenting entry symptoms, asking sequential triage questions following decision-tree logic, mapping answer patterns to a structured diagnosis, and returning a routing recommendation to the orchestrator. Operates entirely as an internal agent — never communicates…
output-agent
Presentation formatting specialist that takes raw analysis results from skill execution or playbook completion and renders them into the appropriate output template for the specified target audience. Selects templates, fills them with analysis data, applies audience-specific tone adjustments, and handles…
playbook-runner-agent
Multi-step execution specialist that runs playbooks by progressing through each step sequentially, invoking the skill-executor-agent for framework application at each step, managing checkpoints and decision points, tracking cumulative progress, and returning compiled results to the orchestrator. Handles branching…
kebab-case-agent-name
Third-person description of this agent's role within the Foreman orchestration layer. Max 1024 characters.