- Home
- Skills
- Agents & Orchestration
- code map skills
More screenshots
Works with the AI Agents you already use
code map skills
Transform codebases and Git diffs into interactive, single-file HTML maps with linked diagrams and narrative tours.
$1.50
code map skills
Example session with this skill installed
Generate a code map for the current PR changes to show how the new authentication middleware impacts the request flow.
- Read your context and instructions
- Compiled the code map
- Generated the UI component
I have analyzed the Git diff for the authentication middleware update. I extracted the changed files, identified the new CheckAuth interceptor, and mapped its interaction with the UserStore and SessionManager.
A new code map has been generated at docs/code-map/auth-middleware-update/auth-update.html.
Code Map: Authentication Middleware Update
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<title>Auth Middleware Review</title>
</head>
<body>
<!-- Engine logic hidden -->
<script>
/* ════════ MAP-DATA-START ════════ */
const META = {
title: "Auth Middleware Update",
sub: "PR #421 - Security Layer",
lang: "en",
stats: { files: 4, inserts: 124, deletes: 12 }
};
const OVER_SRC = `
flowchart LR
Client-->GW[API Gateway]
GW-->Auth[CheckAuth Interceptor]:::newc
Auth-->Sess[SessionManager]
Auth-->Store[UserStore]
classDef newc fill:#e1f5fe,stroke:#01579b
`;
const SEQ_SRC = `
sequenceDiagram
participant C as Client
participant A as CheckAuth
participant S as SessionManager
C->>A: Request with JWT
A->>S: Validate Token
alt Valid
S-->>A: Session Object
A->>A: Inject Context
else Expired
S-->>A: 401 Unauthorized
end
`;
const TREE = [
{ id: "src/auth", text: "auth", children: [
{ id: "mid", text: "middleware.go", status: "MOD", doc: "d1", kw: "CheckAuth" }
]}
];
const DOCS = [
{ id: "d1", title: "CheckAuth Interceptor", content: "Validates incoming JWTs and populates context. Uses a new grace period for token rotation to prevent race conditions during refresh." }
];
const TOUR = [
{ title: "Review Entrypoint", text: "Start here to see the new middleware registration.", tab: "over", tree: "mid" }
];
/* ════════ MAP-DATA-END ════════ */
</script>
</body>
</html>
Next steps
- Open
docs/code-map/auth-middleware-update/auth-update.htmlin your browser. - Click the Start Tour button to follow the request flow through the new logic.
- Use the Modified filter on the right panel to focus only on the changed logic cards.
auth-update.html
TSX · React component
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
Standard READMEs and folder trees fail to explain the execution flow or mental model of complex codebases. Developers on boarding to new projects or reviewing large pull requests often struggle to visualize how components interact across different layers.
What it does
- Standardized Visualization creates a three-panel interactive dashboard featuring project trees, Mermaid diagrams, and deep-dive documentation cards.
- Tri-Directional Linkage synchronizes highlighting across all panels so clicking a tree node instantly focuses its logic in the diagram and its technical details in the documentation.
- Guided Narrative Tours builds a sequence of 6 to 10 stations that walk readers through end-to-end execution paths or implementation steps.
- Multi-Scenario Mapping adapts output for technical teaching, git diff reviews, feature blueprints, or architectural onboarding.
- Zero-Dependency Delivery generates a single, self-contained HTML file that works in any browser without companion folders or local assets.
How it works
- Context Extraction: The agent analyzes your repository, Git diffs, or feature specs to map class hierarchies and logic flows.
- Data Construction: It populates a pre-defined schema with Mermaid sources, project trees, and architectural intent cards.
- Validation: A static analysis script runs to ensure all keywords, documentation links, and diagram syntax are valid.
- HTML Generation: The final map is emitted as a standalone
.htmlfile, optionally inlined with all engines for offline use.
Frameworks & tools
This skill supports Java, Kotlin, TypeScript, JavaScript, Go, Python, and Rust. It utilizes the Mermaid.js engine for rendering flowcharts and sequence diagrams.
Why this beats prompting it yourself
Generic LLMs often produce disconnected diagrams or text descriptions that go out of sync. This skill enforces a hard quality gate via validation scripts and ensures that every visual element is structurally linked to its documentation and code location.
Use cases
- Onboarding new hires by providing a visual walkthrough of the core request lifecycle.
- Visualizing PR reviews to see the blast radius of changes across multiple modules.
- Drafting feature blueprints to map out architecture before writing the first line of code.
- Technical documentation for complex open-source libraries that require more than just Javadoc.
Known limitations
Flowcharts are limited to a maximum subgraph depth of one. Sequence diagrams cannot use raw semicolons in notes or labels.
Flowchart subgraph depth is limited to 1. Sequence diagrams cannot contain raw semicolons in labels. Support is limited to Java, Kotlin, TS, JS, Go, Python, and Rust.
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 any payment.
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
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 7 days ago
Requires
Permissions
Allowed Hosts