Platform-agnostic AI skill: transform any project idea into a complete technical blueprint — stack, schema, API, roadmap, and risk report.
SaferSkills independently audited ai-workflow-architect (Agent Skill) and scored it 83/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 4 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 4 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
description: Transform any project idea into a complete technical blueprint — stack, schema, API, architecture, roadmap, and risk report. argument-hint: <project description> allowed-tools: Read, Write, Bash, WebFetch
Transform a raw project idea into a production-ready technical blueprint.
The skill extracts intent, asks only critical clarifying questions, then generates a structured architecture with tech stack, DB schema, API design, execution roadmap, and LLM-powered risk validation.
Extract from user input:
Always output:
Confidence: <0-100>%
Reason: <one sentence explaining the score>Confidence is based on:
Evaluate whether the extracted information is sufficient to generate a reliable architecture.
Only ask questions that materially change:
Never ask:
Rules:
If confidence is below 80%:
Output only Steps 1 and 2, then stop.
CRITICAL RULE:
When at least one clarification question is asked:
#### Assumption Handling
If the user explicitly says:
then:
Assumptions must be clearly labeled.
Example:
Assumption:
Monetization model = SaaS subscription
Reason:
Most B2B AI tools use subscription pricing for MVP validation.Generate:
#### Tech Stack
For every technology provide:
#### Core Modules
List modules and explain responsibilities.
#### API Structure
List key endpoints grouped by domain.
#### DB Schema
Include:
Generate:
#### MVP Roadmap
Phased roadmap with timelines.
#### Sprint Breakdown
Detailed tasks for Phase 1.
#### Priority Order
Provide a numbered list. For each item include:
Before proceeding to Step 5 verify:
If any answer is YES:
Review the generated architecture for:
Output the top 3 risks. Requirements:
For each risk provide:
Use the following definitions when assigning severity:
#### High
#### Medium
#### Low
Reasonable assumptions are allowed only when they do not materially affect architecture.
If an assumption changes business model, user roles, platform, monetization, or integrations — ask a clarification question instead.
Always label assumptions explicitly.
Before final output verify every checklist item. If any item is missing — fix before responding.
Checklist:
A complete technical blueprint containing:
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.