kortix-executor

Execute third-party API interactions through a unified MCP interface.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/jbreel77888/Agent-AiNorx --skill kortix-executor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kortix-executor
Source: https://github.com/jbreel77888/Agent-AiNorx/tree/main/.kortix/opencode/skills/kortix-executor
Command: npx skills add https://github.com/jbreel77888/Agent-AiNorx --skill kortix-executor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the need for Kortix agents to manually handle raw API tokens, per-app SDKs, or custom integration code when interacting with external third-party systems, reducing security risks and development overhead for cross-system tasks.

Core Features & Use Cases

  • Unified MCP Interface: Access all configured third-party tools (Pipedream, OpenAPI, GraphQL, Slack, Stripe, etc.) via a single stable set of MCP tools, eliminating the need to manage per-app SDKs or raw tokens.
  • Progressive Tool Discovery: Search for tools by natural language intent, inspect full input schemas and risk levels before calling, and use generic request proxies for endpoints without pre-built named actions.
  • Secure Credential Management: All third-party secrets are stored and resolved server-side by the Executor Gateway, with no sensitive credentials ever exposed to the agent sandbox.
  • Use Case Example: Use this Skill to send a Slack notification to your team, create a Stripe charge for a customer, or post a comment on a GitHub pull request without writing any custom API code or handling sensitive credentials.

Quick Start

Ask your Kortix agent to send a Slack message to your #general channel with the text "Sprint planning is scheduled for 2pm today" and it will automatically use the Executor to complete the task securely.

Frequently Asked Questions about kortix-executor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I connect AI agents to external APIs without exposing raw credentials?

To connect AI agents to external APIs without exposing raw credentials, use a secure executor gateway that stores secrets server-side and resolves them dynamically, ensuring sensitive tokens never reach the agent sandbox.

What is the best way to integrate SaaS tools like Slack and Stripe into agent workflows?

The best way to integrate SaaS tools like Slack and Stripe into agent workflows is using a unified MCP interface, which provides a single stable set of tools to access multiple third-party applications without managing per-app SDKs.

How does progressive tool discovery work for SaaS automation?

Progressive tool discovery for SaaS automation works by allowing agents to search for third-party tools using natural language intent, inspect input schemas and risk levels before execution, and use generic request proxies for endpoints lacking pre-built actions.

Do I need to write custom integration code to send a Slack notification through an agent?

No, you do not need to write custom integration code to send a Slack notification through an agent. You can simply ask the agent to perform the task, and the executor will securely handle the API request automatically.

Can I use a single interface to manage payments and internal API queries?

Yes, you can use a single unified MCP interface to manage cross-system tasks like processing Stripe payments, querying internal APIs, and managing SaaS workflows without handling raw API tokens or custom per-service integration code.

What are the limitations of using a unified MCP interface for third-party connectors?

A limitation of using a unified MCP interface for third-party connectors is that endpoints without pre-built named actions require generic request proxies, and all cross-system tasks are subject to enforced user permission policies and auditable integration layers.