# Pathmind > Pathmind (formerly Skymind, the company behind the Deeplearning4j deep-learning library) > was a San Francisco reinforcement-learning company that let simulation engineers train RL > policies against AnyLogic and Python simulations without writing machine-learning code. > Pathmind ceased operating in November 2021; its own final release is titled "Final release > - disabling sign-ups and upgrades". The app, API and marketing hosts no longer serve. This > file is an API Evangelist profile of the surviving public first-party surface. Generated: 2026-08-05 Method: generated (from apis.yml and the artifacts in this repository) Source: https://raw.githubusercontent.com/api-evangelist/skymind/refs/heads/main/apis.yml ## Status - Operational status: defunct as of 2021-11-17 - https://api.pathmind.com — connection refused (probed 2026-08-05) - https://app.pathmind.com — connection refused (probed 2026-08-05) - https://pathmind.com/ — HTTP 403 Cloudflare challenge; origin is a WordPress.com "Active domain connection for this domain not found" page in the 2026-07-18 archive - Surviving first-party surface: the GitHub organization and one PyPI package ## APIs - [Pathmind API](https://github.com/PathmindAI/pathmind-webapp/tree/dev/pathmind-api): Two operations — list the authenticated user's projects (getProjects) and upload an AnyLogic simulation model to create an experiment (uploadAlModel). API-token auth via the X-PM-API-TOKEN header. Base URL https://api.pathmind.com (unreachable). ## Specs - [OpenAPI 3.0.3](https://github.com/PathmindAI/pathmind-webapp/blob/dev/pathmind-api/src/main/resources/openapi.yaml): first-party definition, published in the pathmind-webapp repository - Local copy: openapi/skymind-pathmind-openapi-original.yml - Overlay of API Evangelist enhancements: overlays/skymind-pathmind-overlay.yaml ## Artifacts - Authentication profile: authentication/skymind-authentication.yml - API conventions: conventions/skymind-conventions.yml - Error catalog: errors/skymind-problem-types.yml - Data model: data-model/skymind-data-model.yml - Conformance: conformance/skymind-conformance.yml - Lifecycle: lifecycle/skymind-lifecycle.yml - Changelog: changelog/skymind-changelog.yml - Packages / SDKs: packages/skymind-packages.yml - Well-known probe index: well-known/skymind-well-known.yml - Domain security probe: security/skymind-domain-security.yml - Agent skills: skills/_index.yml ## SDKs - [pathmind on PyPI](https://pypi.org/project/pathmind/) — first-party Python simulation API, v0.4 (2021-11-04), MIT. `pip install pathmind`. Implement the Simulation interface, then call .train() to train a policy on Pathmind's cloud. ## Source code - [PathmindAI on GitHub](https://github.com/PathmindAI) - [pathmind-webapp](https://github.com/PathmindAI/pathmind-webapp) — the Pathmind web application, including the pathmind-api module that publishes the OpenAPI - [pathmind-api](https://github.com/PathmindAI/pathmind-api) — the Python simulation API - [nativerl](https://github.com/PathmindAI/nativerl) — apply RL to AnyLogic or Python simulations - [policy-server](https://github.com/PathmindAI/policy-server) — FastAPI/Ray policy serving, which generates its own OpenAPI at runtime from a per-model schema.yaml ## Not published - No MCP server, no A2A agent card, no AsyncAPI, no webhooks, no gRPC/protobuf - No OAuth (API-key auth only), no scopes - No status page, no deprecation policy, no SLA, no first-party security.txt - No sandbox, no CLI, no embedded components, no Postman collection