- Home
- Skills
- Code Quality & Review
- FlowLens Code Map
More screenshots
Works with the AI Agents you already use
FlowLens Code Map
Turn a JavaScript, TypeScript, or Python repository into an interactive offline architecture map with source evidence and simulated flow journeys.
Free
FlowLens Code Map
Example session with this skill installed
Use the FlowLens plugin/CLI to map the sample repository. Trace the checkout request through its API route and handler, show data reads and writes, include source file:line evidence, and explain which connections are inferred or unresolved. Generate the offline HTML map and graph JSON.
- Read your context and instructions
- Compiled the flowlens code map
Example based on the included FlowLens shop-app test fixture, not a scan of a buyer’s repository:
CheckoutPage.submitcallscreateOrder(observed).- The frontend request
POST /api/ordersis linked to the FastAPI route by matching method and path (inferred; route evidence:api/app/routes/orders.py:10). - That route handles
create_order, which callsplace_order. save_orderwrites toorders;stock_levelreadscache.POST /api/feedbackremains unresolved because no matching route was scanned.
The report shows evidence paths and lines for source-backed connections. Observed links are solid, inferred links are dashed with the matching reason, and unresolved targets stay unknown. Journeys are simulated from source order, not runtime traces. The scan outputs an offline interactive HTML report and a machine-readable JSON graph.
Connects securely to your tools. The creator never sees your data.
About this skill
FlowLens is a local-first code-mapping CLI and agent plugin. This skill pack gives compatible coding agents two workflows: map a repository and configure scan scope. FlowLens statically analyzes JavaScript, TypeScript, and Python, then creates an interactive offline HTML map and a JSON graph with file and line evidence for API requests and routes, data reads and writes, module imports, and agent handoffs.
FlowLens labels connections as observed, inferred, or unresolved. Inferred links include the rule that connected the source facts; runtime-decided targets remain unresolved. Replays are simulated from source order, not recorded runtime traces.
Requires the FlowLens CLI or plugin to be installed already, plus Node.js 20 or later. Python 3.8 or later is needed to analyze Python files. Scans work locally and do not need network access. This listing provides the agent workflows; it is not an installer or replacement for the FlowLens plugin.
Requires pre-installed FlowLens CLI, Node.js 20+, and Python 3.8+. Analysis is static; does not reflect runtime execution or actual network traffic.
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
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 9 days ago
- Free with an Agensi account
Requires
Permissions
This skill pack wraps the FlowLens plugin workflows; it does not install or replace the plugin. Scans are static and local. Generated maps are written under .flowlens/ by default.