brekkylab/backlot

A local emulator for enterprise SaaS APIs — Slack, Gmail, Google Drive, GitHub, Jira, Notion, S3 and more — with the real response shapes, pagination, auth and per-document ACLs, over a corpus you supply.

What it solves

Backlot is a local emulator for enterprise SaaS APIs (such as Slack, Gmail, GitHub, and Jira). It allows developers to build and test integrations without needing actual vendor accounts, OAuth approvals, secrets in CI, or an active network connection. It solves the problem of "mocking" APIs by providing vendor-shaped responses, pagination, and authentication over a deterministic corpus of data that the developer controls.

How it works

Backlot runs a local server that mimics the request and response contracts of official vendor SDKs. It uses a single SQLite corpus to store facts (messages, files, issues, etc.), from which it derives stable IDs, users, and tokens. By changing the base URL of an official SDK to point to thelot local server, the application code remains identical to production code while interacting with the emulator.

Who it’s for

  • Integration Developers: Those building or upgrading connections to enterprise tools.
  • QA and CI/CD Engineers: Those needing deterministic, secret-free tests that don't flake.
  • AI Agent and RAG Developers: Those evaluating pipelines using a consistent dataset to ensure that changes in scores are due to code changes, not data drift.
  • Bug Fixers: Developers reproducing issues caused by specific data shapes in a user's workspace.

Highlights

  • Broad Support: Emulates a documented subset of APIs for Slack, Gmail, Google Drive, GitHub, Jira, Confluence, Notion, Linear, HubSpot, Fireflies, and Amazon S3.
  • Deterministic Data: Supports importing custom test worlds via JSONL files to ensure consistent results across environments.
  • Identity Modeling: Generates deterministic credentials and implements per-document access control lists (ACLs) to test authentication and permissions.
  • Agent-Ready: Includes an MCP (Model Context Protocol) server to allow coding agents to interact with the mocked environment instantly.

Related

  • Project
    drmingler/smart-llm-loader

    A Python package that converts documents and scanned images into clean, structured markdown chunks using OCR and LLM-powered semantic chunking for RAG systems.

  • Project
    dengyier/OpenWorkProof

    An open protocol for AI Agent work contracts and verifiable execution, ensuring that agent actions are authorized, evidenced, and formally accepted by humans.

  • Project
    mohmdelsayed/streaming-drl

    An implementation of the Stream-X framework for deep reinforcement learning, enabling agents to learn directly from a stream of experience without needing a replay buffer.

  • Project
    Cognitive-AI-Systems/pogema

    A flexible grid-based environment for Partially-Observable Multi-Agent Pathfinding (PO-MAPF), designed for testing decentralized navigation and collision avoidance in multi-agent systems.