- Home
- Skills
- Mobile Development
- flowlint
Works with the AI Agents you already use
flowlint
Extract and audit UX flow diagrams directly from web and mobile source code with file-level traceability.
Free
flowlint
Example session with this skill installed
Analyze the onboarding flow in this React Native project and generate a draw.io diagram. Focus on the signup guards and external OAuth steps.
- Read your context and instructions
- Compiled the flowlint
- Generated the document
I have analyzed the onboarding logic in the src/navigation/AuthStack.tsx and src/screens/auth/ directories. I found a complex signup guard involving email verification and a third-party OAuth redirect that lacked a modeled failure path.
UX Flow Audit: Onboarding
| Step | Type | Source | Finding |
|---|---|---|---|
| Splash Screen | Entry | Splash.tsx:12 | - |
| Email Input | Interaction | SignUp.tsx:45 | UXF-NOERR-0A7D: No validation error UI |
| OAuth Hand-off | Guard | AuthService.ts:88 | High: No handler for user cancellation |
| Verification OTP | Wait | Verify.tsx:22 | UXF-MISS-0B22: Missing "Resend Code" link |
The full analysis and diagram files have been generated:
docs/ux-flows/onboarding.flow.json: The source IR file.docs/ux-flows/onboarding.drawio: Editable diagram (open in app.diagrams.net).docs/ux-flows/onboarding.md: Full audit report with embedded Mermaid flow.
Next steps
- Review the
UXF-NOERR-0A7Dfinding regarding the missing OAuth cancellation path inAuthService.ts. - Open the
.drawiofile to visualize the proposed recovery branch. - Run
flowlint ignorefor any intentional friction points identified.
AuthStack.tsx
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Inheriting a codebase often means flying blind. Design docs rarely match what was actually shipped, leaving developers to guess how error states, empty screens, and edge cases are handled. This skill extracts the ground truth from your source code, generating verifiable UX maps and audits rooted in your actual implementation.
What it does
- Source-anchored mapping creates flow diagrams where every node links to a specific file and line number.
- Automated UX auditing identifies friction points, unhandled errors, and dead ends directly from the logic.
- Multi-format rendering exports flows to editable
.drawiofiles, Mermaid diagrams, and Markdown reports. - Diff analysis compares current flows against proposed changes to measure the delta in steps or friction.
- CI verification ensures your documentation stays in sync with code via automated staleness checks.
How it works
- Inventory the stack by scanning for web, mobile, or native frameworks like Next.js, React Native, or Flutter.
- Trace the logic from entry points to goals, documenting guards, network calls, and all failure branches.
- Generate IR (Intermediate Representation) files in JSON format to maintain a deterministic source of truth.
- Render deliverables using the included Python utility to produce diagrams and finding reports.
Frameworks & tools
Supports Next.js, React, React Router, React Native, Expo, Flutter, SwiftUI, UIKit, and Jetpack Compose. Requires Python 3.8+ for the rendering engine.
Why this beats prompting it yourself
Generic prompts often hallucinate UI steps or miss obscure code branches like permission denials and timeout recovery. This skill follows a strict IR schema and use code-level evidence, ensuring the diagrams represent reality rather than a guess.
Use cases
- Codebase handovers to quickly visualize how complex authentication or payment flows actually work.
- Refactoring plans to visualize the "before and after" of a proposed UX change before writing a line of code.
- QA automation to identify every branch and error state that requires a test case.
- Audit documentation for compliance or product reviews that stay updated via your CI pipeline.
Known limitations
Cannot design non-existent flows or analyze visual design elements. Dynamic runtime routing may result in incomplete maps. Requires manual identification of specific flows to be mapped.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Mobile Development.
Cannot analyze flows not yet written or visual design. Dynamic runtime routing may cause incomplete maps. Requires Python 3.8+ for rendering engine.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
10 uses
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 25 days ago
- Free with an Agensi account
- Software
Python
Git