Hermes Atlas
Developer tools & SDKs · works with Hermes Agent

AI Native Lang (AINL)

sbhooley/ainativelang

Compact graph-based language for deterministic AI agent workflows, installable into Hermes over MCP

In short

AI Native Lang (AINL) is a compact, graph-canonical programming language and toolchain for writing AI workflows that are validated and run deterministically. An MCP setup command adds it to hosts including Hermes, OpenClaw and Claude Code.

What AI Native Lang (AINL) does

AINL targets teams whose agents author orchestration code, or whose recurring jobs re-prompt an LLM on every run just to decide routing. Workflows are written in a compact graph form, validated at compile time, and executed with state, memory and tool use, with the stated aim of lowering reliance on long prompt loops. The README lists emitting one workflow source to LangGraph, Temporal and FastAPI as a use case, along with tamper-evident execution traces for compliance audits.

A single command, pipx install 'ainativelang[mcp]' && ainl setup --auto, detects installed hosts, including Claude Code, Cursor, Cline, Codex, Claude Desktop, OpenClaw, Hermes and ArmaraOS, and merges an MCP server entry into each config with an atomic write and a timestamped backup. The setup is idempotent, and ainl doctor verifies it. The README is candid about who should skip AINL, such as teams whose CI already catches runner bugs, and it links ArmaraOS, a desktop agent OS built on AINL.

Key features

  • Compact graph-canonical language with compile-time validation
  • One-command setup that merges an MCP entry into detected hosts
  • Atomic config writes with timestamped backups and an ainl doctor check
  • Deterministic execution intended to cut repeated LLM routing calls
  • ainl setup --print-config for hosts that are not auto-detected

When to use it

  • Replacing LLM-routed scheduled monitors and digests with validated workflows
  • Letting a Hermes agent author and run workflows over MCP
  • Producing audit-friendly execution traces for compliance needs

Who it is for: Teams building multi-step AI workflows that need repeatable execution, validation and control.

How it fits with Hermes Agent

Installs into Hermes among many MCP hosts: ainl setup --auto detects Hermes and merges an MCP server entry into its config.

How to install AI Native Lang (AINL)

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

pipx install 'ainativelang[mcp]' && ainl setup --auto
python3 -m pip install --user 'ainativelang[mcp]' && ainl setup --auto

Requirements: pipx, or Python with pip for the fallback command

FAQ

What is AI Native Lang (AINL)?

AI Native Lang (AINL) is a compact, graph-canonical programming system for AI workflows. It compiles and validates multi-step workflows with state, memory and tool use so they run repeatably instead of relying on long prompt loops.

Does AI Native Lang (AINL) work with Hermes Agent?

Yes. The README says AINL installs directly into OpenClaw, ZeroClaw, Hermes, Claude Code and any MCP-compatible agent, and that ainl setup --auto detects Hermes and adds the MCP server entry.

How do I install AI Native Lang (AINL)?

Run pipx install 'ainativelang[mcp]' && ainl setup --auto. If pipx is unavailable, use python3 -m pip install --user 'ainativelang[mcp]' && ainl setup --auto, and ainl doctor verifies the result.

Similar dev tools for Hermes Agent

All dev tools

Related guides: How to install Hermes Agent · Run multiple Hermes agents with profiles