blaktron/app-architecture-designer
interactivev0.2.0
Interviews you about an app you want, then writes its high-level architecture — components, data, decisions, build order.
Declared capabilities
data: files
Capabilities are the author's declaration and the basis for policy and scanning. The registry cannot enforce them at run time. Your runtime should.
MCP servers & tools
agentry run serves exactly the tools this contract declares, through a per-run MCP gateway. An undeclared capability is a tool that does not exist at run time.
Tools the runner grants
| Manifest declaration | Tool(s) granted |
|---|---|
| (always) | write_report — reports → the run's output folder |
| (always) | list_inputs, read_input, search_inputs — operator-provided input files, read-only |
| (always) | scratch_list, scratch_read, scratch_write, scratch_search — the run's own scratch folder at /scratch, private to the run and deleted at its end; the runner's grant, not a line of the contract |
| (always) | export_pdf — a Markdown report as a PDF in reports/, laid out in your report design; the runner's grant, not a line of the contract |
| (always) | grep, glob — search file contents by regular expression and list files by path pattern (ripgrep), one area a call: the inputs, attached knowledge, /scratch; the runner's grant, not a line of the contract |
MCP servers involved
- QuickAgent Tools — the runner's companion server, served through a per-run Docker MCP Gateway; its tool surface is exactly the table above.
- Operator-cataloged MCP servers may be attached for the declared dataAccess category (files); undeclared categories are never wired in.
In an embedded install
Installed into an existing coding agent (agentry install), the same contract maps to the host's built-in tools — enforcement there is the host's. Allowed tools: none declared (host defaults apply).
Install
agentry install blaktron/app-architecture-designer@0.2.0git clone https://secureagentry.net/blaktron/app-architecture-designer.gitgit clone ssh://git@git.secureagentry.net/blaktron/app-architecture-designer.gitclaude-skill is this agent's default target, so the flag is not needed.
agent.yaml
2.2 kBBack to files
schemaVersion: "2"
name: app-architecture-designer
namespace: blaktron
modes: [interactive]
version: 0.2.0
summary: Interviews you about an app you want, then writes its high-level architecture — components, data, decisions, build order.
license: MIT
tags: [architecture, software-design, app-planning, requirements, interview]
entrypoints:
instruction: AGENT.md
intake:
- id: brief
prompt: What app do you want to design? A sentence or two is enough; you can also just say it in the chat.
guide:
overview: 'Runs a short interview about the app you have in mind — who uses it, the core flows, the data, the platform, your team and launch window — then designs its high-level architecture. It hands back architecture.md: the components and what each owns, the data model and where the source of truth sits, authentication, the key request paths, the decisions with the alternatives you did not take and when to switch, and a milestone order that starts with a thin end-to-end slice. Anything you leave unanswered is written down as a named assumption with its default, so you can correct it instead of starting over.'
prepare: [A sentence on what the app does and who it is for., Who builds it and what they already work in — this decides the stack., "A rough launch window, and a budget ceiling if you have one.", 'Anything fixed: a platform, a system it must connect to, a region the data must stay in.']
files: 'Optional: an existing brief, PRD, feature list, or notes on a stack already in place to read before the interview.'
task: e.g. A booking app for dog groomers — customers book and pay, groomers run their calendar.
interactive:
opening: Tell me the app in a sentence or two — what it does and who it's for. I'll ask a few short questions, then write you an architecture you can start building from.
runtime:
interfaces: [claude-skill, agents-md]
capabilities:
network:
egress: none
filesystem:
read: none
write: none
exec: none
dataAccess: [files]
knowledge:
accepts:
prompt: 'Optional: your team''s conventions, or the docs of a system the app must connect to, as a knowledge pack.'
publish:
include: [agent.yaml, AGENT.md, README.md, LICENSE]