Enforce coding-agent protocols mechanically with Harmonist
Drop a multi-agent protocol pack into a project so coding assistants must satisfy reviewer, memory, and supply-chain gates before code-changing work completes.
npx skills add agentskillexchange/skills --skill enforce-coding-agent-protocols-mechanically-with-harmonist
integration-prompt.md into a supported agent session or run python3 harmonist/agents/scripts/integrate.py --pack harmonist --project .; verify the integration with the repository's verify_integration.py workflow.Use Harmonist when prompt-only coding-agent rules are not enough. The operator integrates the pack into a project, lets the assistant generate project-specific rules, installs hooks and structured memory, dispatches specialist agents, and verifies that code-changing work satisfies required protocol gates before completion. The agent workflow is to read the integration prompt or run the integration script, install the generated `AGENTS.md` and hook configuration, route work through the appropriate specialist agents, then run verification so reviewer, memory, and supply-chain requirements are checked mechanically. The boundary is hook-backed protocol enforcement and supply-chain validation for coding-agent workflows, not a generic agent catalogue or IDE plugin. Multi-Framework is defensible because upstream explicitly targets several coding assistants and agent runtimes.
What this skill actually does
Inputs and prerequisites: Python 3.9+, Git, a supported coding assistant such as Cursor, Claude Code, Copilot, Windsurf, Aider, OpenCode, OpenClaw, or Gemini CLI.
Setup notes: Clone the Harmonist repository, then either paste integration-prompt.md into a supported agent session or run python3 harmonist/agents/scripts/integrate.py –pack harmonist –project . ; verify the integration with the repository’s verify_integration.py workflow.
Source and verification boundary: use https://github.com/GammaLabTechnologies/harmonist as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.