agenfk

Enforce structured AI-assisted software engineering workflows with MCP gatekeepers.

64|8|Updated Feb 20, 2026
One-click install
npx skills add https://github.com/cglab-public/agenfk --skill agenfk
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agenfk
Source: https://github.com/cglab-public/agenfk/tree/main
Command: npx skills add https://github.com/cglab-public/agenfk --skill agenfk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI-assisted development often lacks governance, consistency, and traceability. AgenFK provides a mechanical, auditable framework to enforce a high-quality engineering workflow for AI agents.

Core Features & Use Cases

  • Mandatory workflow enforcement across editor integrations (CLAUDE, Opencode, Cursor, Codex) via MCP gatekeepers and pre-tool hooks.
  • Dual operating modes: Standard (single-agent) and Deep (multi-agent orchestration) with plan decomposition and automated handoff.
  • Rich verifications: per-item build/test gates, 80% test-coverage enforcement, and a flow-aware progression system.
  • Visual, real-time Kanban board with multi-project support and GitHub Issues sync for cross-team collaboration.
  • Extensive telemetry and metrics for token usage, cycle time, and flow health.

Quick Start

Start by initializing AgenFK in your editor to establish the project context and framework rules.

Frequently Asked Questions about agenfk

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

FAQPage Schema
How do I enforce a structured AI workflow for software engineering?

You can enforce a structured AI workflow using MCP gatekeepers and pre-tool hooks that mandate compliance and traceability across multiple editor integrations before development proceeds.

How do I track progress and visualize tasks for AI-assisted development?

You can track AI-assisted development progress through a real-time Kanban board that supports multiple projects and synchronizes with GitHub Issues to facilitate cross-team collaboration and visual task management.

Can I use AI workflow governance tools across different code editors?

Yes, AI workflow governance integrates directly with editors like CLAUDE, Opencode, Cursor, and Codex by applying MCP tools and pre-tool hooks to maintain consistent engineering standards.

How do I measure token usage and cycle time in AI-assisted coding?

You can measure token usage, cycle time, and flow health by leveraging built-in telemetry and metrics tools that log token consumption and validate progress throughout the development lifecycle.

What is the difference between Standard and Deep modes in AI workflow orchestration?

Standard mode manages single-agent operations, while Deep mode handles multi-agent orchestration with plan decomposition and automated handoff to coordinate complex AI-assisted engineering tasks.

How do I ensure test coverage and build compliance in AI-generated code?

You can ensure build compliance by applying per-item build and test gates alongside an 80% test-coverage enforcement rule, using a flow-aware progression system to block non-compliant steps.