- Home
- Skills
- APIs & Backend
- HAR to API Map
Works with the AI Agents you already use
HAR to API Map
Convert browser .har captures into structured API maps, JSON schemas, and integration docs without leaking credentials.
$19
HAR to API Map
Example session with this skill installed
Analyze api-map.json from my capture of the dashboard and generate a markdown API map with the auth story and observed endpoints.
- Read your context and instructions
- Compiled the har to api
I have analyzed the API map generated from your dashboard capture. The capture includes 14 distinct endpoint groups across the primary host.
API Map: Dashboard Services
Auth Story
Authentication is handled via the Authorization header using the Bearer scheme. The capture also indicates the presence of a cf-access-token cookie for session persistence. All integration attempts should store these values in environment variables; do not hardcode the observed values.
Observed Endpoints
| Method | Endpoint | Status Codes | Auth | JSON Shape (Keys Only) |
|---|---|---|---|---|
GET | /api/v1/profile | 200 | Bearer | {id: int, email: string, settings: object} |
GET | /api/v1/projects/:id | 200, 404 | Bearer | {uuid: string, name: string, owner_id: int} |
POST | /api/v1/projects | 201 | Bearer | Req: {name: string} Res: {uuid: string} |
PATCH | /api/v1/settings | 200 | Bearer | Req: {theme: string, notifications: bool} |
Not covered / to verify
- Delete Operations: No
DELETErequests were observed in this capture.
Pagination: The /api/v1/projects list was only captured with a single page; limit and offset parameters were not exercised.
- Terms of Service: Review the site's ToS regarding automated access before finalizing the integration.
Next steps
- Run a targeted capture of the project deletion flow to map the
DELETEendpoint shape. - Define
API_BEARER_TOKENin your.envfile for local testing. - Verify if the
cf-access-tokencookie is required for server-to-server calls or only for browser sessions.
Connects securely to your tools. The creator never sees your data.
About this skill
Reverse-engineering a private API from browser traffic is tedious and prone to security leaks. Manual extraction of JSON shapes and header requirements often results in broken integrations or accidental credential exposure.
This skill automates the transformation of .har browser captures into structured API maps. It identifies endpoint patterns, normalizes dynamic paths (like /user/123 to /user/:id), and extracts request/response JSON schemas without ever touching sensitive data values. It provides a clean foundation for building clients, mocks, or documentation based on observed traffic.
What it does
- Path normalization collapses unique URLs into logical endpoint groups using pattern matching for IDs and UUIDs.
- Schema extraction generates structural JSON shapes for request and response bodies, identifying keys and data types only.
- Auth detection identifies authentication schemes such as Bearer tokens, API keys, and session cookies from header presence.
- Parameter mapping catalogs query string parameters and form-urlencoded fields along with their observed scalar types.
- Traffic analysis provides endpoint statistics including request counts, status code distribution, and average latency.
How it works
- Analyze capture by running the provided Node script against a local
.harfile to generateapi-map.json. - Review map to filter out third-party telemetry, analytics, and static asset noise from the target API endpoints.
- Generate deliverable such as a TypeScript client, an OpenAPI spec draft, or integration documentation based on the mapped shapes.
Frameworks & tools
Node.js 22+, DevTools HAR exports, and JSON-based schema modeling.
Why this beats prompting it yourself
Generic prompts often hallucinate API structures or, worse, try to process raw HAR files containing live session tokens. This skill uses a local, zero-dependency script to strip all values before the LLM sees the data, ensuring security while providing perfect structural accuracy that manual prompting cannot match.
Use cases
- Scaffold a typed SDK for a web application that lacks an official public API.
- Document internal microservice communication for legacy system migration.
- Create realistic mock servers by capturing production-like request and response shapes.
- Identify missing error handling by mapping observed 4xx and 5xx response bodies.
Known limitations
Does not capture endpoints or parameters not exercised during the browser session. Requires Node 22+ for the local mapping script. Cannot verify API rate limits or exhaustive field requirements.
Requires Node 22+. Only maps endpoints explicitly triggered during recording. Cannot infer hidden parameters or undocumented fields not present in the capture.
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 21 days ago
Requires
Permissions
Reads a local .har file and writes api-map.md / api-map.json next to it. No network calls, no credential values ever emitted.