awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
UsefulSoftwareCo avatar

UsefulSoftwareCo/executor

0
View on GitHub↗

Executor

Executor is an integration layer that connects AI agents to external apps and APIs through a single shared endpoint. Agents get a search tool and a code-running tool, so they can write one short program that calls tools from many apps within a single task. That program runs in a sandboxed environment, and the endpoint is reachable by any client that speaks the Model Context Protocol, such as coding assistants and chat apps.

Credentials and approvals are handled outside the agent. An API key or sign-in is saved once and attached to outgoing calls inside the app layer, so the agent never sees the raw secret. Each tool also carries its own approval rule: ask first, always allow, never allow, or check the input for risk. Approvals can be routed through the agent, a client prompt, or a browser page.

Apps can be created from OpenAPI specs, GraphQL schemas, remote tool servers, or hand-written API calls, which are wrapped as small typed programs. They can be installed from a catalog or added by a remote server's web address, and an agent can write and deploy new versions that are published as immutable versions and can be rolled back.

Executor can run as a hosted service, a local background process with local data, or a single container with an embedded database, all serving the same apps. Groups of people can share connected accounts through browser sign-in in hosted or self-hosted setups.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI
executor.sh
↗

Features

  • AI Tool Integration Layers - Connects AI agents to external apps, APIs and tool servers through one consistent interface, with credential handling and approval policies outside the agent.
  • MCP Tool Gateways - Exposes every connected app and tool through one shared endpoint reachable by any Model Context Protocol client, such as coding assistants and chat apps.
  • Agent Code Execution Across Tools - Give agents a search tool and an execute tool so they write one short program that calls many apps' tools within a single task.
  • Tool Execution Approvals - Decides per tool whether to ask first, always allow, never allow, or check input risk before an agent runs it.
  • AI Agent Tool Integrations - Connects AI agents to many external apps and APIs, including OpenAPI, MCP, and GraphQL tools, from one place within a task.
  • API Spec to Agent Tool Converters - Wraps OpenAPI specs, GraphQL schemas, remote tool servers, and hand-written API calls as small typed programs that agents can call directly.
  • MCP Gateways - Exposes connected apps and tools to any Model Context Protocol client through a single shared endpoint.
  • Multi-Tool Program Execution - Gives agents a search tool and a code-running tool so one short program can call tools from many apps in a single task.
  • Tool Approval Rules - Lets each tool carry its own rule to ask first, always allow, never allow, or check input risk, with approvals via agent, client, or browser.
  • Code Execution Sandboxes - Runs agent-written programs in an isolated sandbox with restricted access, keeping untrusted code away from the host and secrets.
  • Agent Program Sandboxes - Runs short agent-written programs in an isolated sandbox so they can chain calls across many tools without touching the host or secrets.
  • Agent-Authored Tool Apps - Installs apps from a catalog, adds remote servers by web address, and lets agents write and deploy new rollback-able app versions.
  • Connected App Tools - Exposes all connected apps and tools to any compatible client, including coding assistants and chat apps, via one shared endpoint.
  • Human-in-the-Loop Workflows - Requires a person's confirmation for risky agent actions before they run, via the agent, a client prompt, or a browser page.
  • Shared Team Account Access - Organizes people into groups who share connected accounts through browser sign-in, in hosted or self-hosted setups.
  • OpenAPI-to-Tool Converters - Turns OpenAPI specs, GraphQL schemas, remote tool servers and hand-written API calls into small typed tools that agents can call.
  • MCP Endpoint Grouping - Exposes every connected app and tool through one shared MCP endpoint that coding assistants and chat clients can reach.
  • Cloud or Self-Hosted Deployments - Runs as a hosted service, a local background process, or a single container with an embedded database, all serving the same apps.
  • Stored Credential Retrievals - Saves an API key or sign-in once and reuses it across apps, attaching it in the app layer so agents never see it.
  • Credential Injection - Attaches saved API keys and sign-ins to outgoing tool calls inside the app layer so agents never see raw secrets.
  • Self-Hosted Vaults - Stores API keys and sign-ins once, in a hosted or self-hosted setup, and injects them into outgoing calls so agents never see the secrets.
4,363 stars·403 forks·TypeScript·MIT·4 views

Star history

Star history chart for usefulsoftwareco/executorStar history chart for usefulsoftwareco/executor

How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

Projects sharing features with Executor

These projects share indexed features with Executor. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
  • zenml-io/zenmlzenml-io avatar

    zenml-io/zenml

    5,637View on GitHub↗

    ZenML is an orchestration platform designed for building, deploying, and monitoring reproducible machine learning pipelines and agentic workflows. It provides a unified framework that manages the entire lifecycle of machine learning assets, from data processing and model training to the deployment of persistent inference services. By decoupling pipeline logic from underlying compute and storage, the platform enables teams to transition workflows seamlessly from local development environments to production-grade cloud infrastructure. The platform distinguishes itself through a service-oriented

    Pythonagentopsagentsai
    View on GitHub↗5,637
  • withcoral/coralwithcoral avatar

    withcoral/coral

    4,937View on GitHub↗

    Coral is a self-hostable execution layer for AI agents. Instead of loading every tool definition into a prompt, an agent searches a typed catalog for the tools it needs and then writes a short program that calls several connected services in one step, returning only the fields the answer requires. Those programs run in an isolated sandbox with no secrets, no network access, and no imports, and service tokens are kept in a central store and attached to outbound calls on the server side, so the agent never sees a credential. Coral also acts as a governance layer. Each tool can be set to Allow,

    Rust
    View on GitHub↗4,937
  • prefecthq/fastmcpPrefectHQ avatar

    PrefectHQ/fastmcp

    22,994View on GitHub↗

    FastMCP is a Python framework designed for building servers that expose functions, resources, and prompts to AI models using the Model Context Protocol. It simplifies the development process by automatically deriving tool metadata, input schemas, and documentation directly from Python function signatures and type hints. The framework provides a unified container for managing these components, allowing developers to build modular applications that integrate seamlessly with AI assistants. The project distinguishes itself through its support for interactive, server-defined user interface compone

    Pythonagentsfastmcpllms
    View on GitHub↗22,994
  • openai/openai-agents-pythonopenai avatar

    openai/openai-agents-python

    27,191View on GitHub↗

    This project is a Python framework for building autonomous, event-driven agent systems. It provides a unified runtime for orchestrating multi-agent workflows, managing persistent conversation state, and executing code within secure, isolated sandbox environments. The framework is designed to handle complex task delegation, allowing agents to invoke other agents as tools while maintaining context across multi-turn interactions. The framework distinguishes itself through its deep integration with the Model Context Protocol, enabling agents to connect to external data sources and remote services

    Pythonagentsaiframework
    View on GitHub↗27,191
Compare all 30 related projects→

Frequently asked questions

What does usefulsoftwareco/executor do?

Executor is an integration layer that connects AI agents to external apps and APIs through a single shared endpoint. Agents get a search tool and a code-running tool, so they can write one short program that calls tools from many apps within a single task. That program runs in a sandboxed environment, and the endpoint is reachable by any client that speaks the Model Context Protocol, such as coding assistants and chat apps.

What are the main features of usefulsoftwareco/executor?

The main features of usefulsoftwareco/executor are: AI Tool Integration Layers, MCP Tool Gateways, Agent Code Execution Across Tools, Tool Execution Approvals, AI Agent Tool Integrations, API Spec to Agent Tool Converters, MCP Gateways, Multi-Tool Program Execution.

Which projects share features with usefulsoftwareco/executor?

Projects with overlapping indexed features include: zenml-io/zenml — ZenML is an orchestration platform designed for building, deploying, and monitoring reproducible machine learning… withcoral/coral — Coral is a self-hostable execution layer for AI agents. Instead of loading every tool definition into a prompt, an… prefecthq/fastmcp — FastMCP is a Python framework designed for building servers that expose functions, resources, and prompts to AI models… openai/openai-agents-python — This project is a Python framework for building autonomous, event-driven agent systems. It provides a unified runtime… elie222/rakazo — Rakazo is a self-hosted, open-source platform of persistent AI bots. Each bot keeps its own conversations, memory,… vrsen/agency-swarm — Agency Swarm is a multi-agent orchestration framework and development kit designed to coordinate specialized AI agents…