hitl-test

Execute human-in-the-loop test scenarios against a live agent-intercom server.

2|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/softwaresalt/agent-intercom --skill hitl-test
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hitl-test
Source: https://github.com/softwaresalt/agent-intercom/tree/main/.github/skills/hitl-test
Command: npx skills add https://github.com/softwaresalt/agent-intercom --skill hitl-test

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a robust framework for manually testing the agent-intercom's core functionality, ensuring its Slack integration, approval workflows, and operational modes function as expected in real-world scenarios.

Core Features & Use Cases

  • End-to-End Testing: Validates the entire human-in-the-loop (HITL) process from agent action to Slack notification and operator response.
  • Mode Verification: Tests both MCP and ACP operational modes, including session management, threading, and workspace routing.
  • Use Case: Before deploying agent-intercom to production, run this Skill to simulate user interactions and verify that code changes are correctly presented for approval, sessions are managed properly, and all notifications appear as intended in Slack.

Quick Start

Run the hitl-test skill to execute the HITL tests for the agent-intercom server.

Frequently Asked Questions about hitl-test

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

FAQPage Schema
How do I test human-in-the-loop approval workflows in Slack before deploying?

Verify Slack integration for agent-intercom by executing end-to-end HITL tests that validate notifications, approvals, and operational modes. This confirms sessions are managed properly and all notifications appear as intended in the Slack workspace.

How do I verify Slack integration for agent-intercom operational modes?

Verify Slack integration by executing end-to-end tests that validate notifications, approvals, and operational modes. This confirms sessions are managed properly and all notifications appear as intended in the Slack workspace.

What human-in-the-loop scenarios can I validate for MCP and ACP sessions?

Human-in-the-loop scenarios for MCP and ACP sessions include validating ping, broadcast, approval and rejection, modal capture, steering queue, task inbox, session management, and audit logging to ensure correct operational mode functionality.

Does testing agent-intercom Slack integration require active operator participation?

Testing agent-intercom Slack integration requires active operator participation via Slack for approvals and confirmations. An operator must manually respond to notifications during the test scenarios to complete the validation loop.

What is the best way to validate agent-intercom audit logging and task inbox routing?

The best way to validate agent-intercom audit logging and task inbox routing is by running targeted HITL test scenarios against a live server. This verifies that workspace routing, threading, and audit logs function as expected.

Why does my agent-intercom Slack approval test fail during broadcast and modal capture?

Agent-intercom Slack approval tests may fail during broadcast and modal capture if the live server is not properly configured for MCP and ACP modes. Running the test scenarios validates these specific paths to identify configuration issues.