Menu
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/sbhooley/ainativelang --skill ainlThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Based on SOC occupation classification
Showing SKILL.md
| name | AINL |
| description | Deterministic compiled graphs for OpenClaw (importer, runtime, MCP tools) |
| install | ./install.sh |
| install_alt | pip install 'ainativelang[mcp]' && ainl install-mcp --host openclaw |
| commands | ["ainl import <source> — convert Markdown / ecosystem packs into `.ainl` graphs","ainl compile <file.ainl> — compile and validate deterministic graph IR","ainl run <file.ainl> — run compiled workflow (when your graph uses the runtime path)","~/.openclaw/bin/ainl-run <file.ainl> — compile-then-run wrapper after install-mcp --host openclaw (add ~/.openclaw/bin to PATH)"] |
| tags | ["ainl","graphs","mcp","openclaw"] |
| metadata | {"openclaw":{"kind":"skill","upstream":"https://github.com/sbhooley/ainativelang","package_pypi":"ainativelang","mcp_server_key":"ainl","config_file":"~/.openclaw/openclaw.json","workspace_default":"~/.openclaw/workspace","onboarding_note":"OpenClaw expects `npm install` + `openclaw onboard` for the host CLI; this skill adds AINL (pip) + MCP + ainl-run.","install_channels":["ClawHub (when listed)","Manual copy to ~/.openclaw/skills or <workspace>/skills"]}} |
AINL is an AI-native language that compiles deterministic graphs. In OpenClaw, use it to import workflows, compile them, run them, and expose ainl-mcp tools to the agent via ~/.openclaw/openclaw.json.
Package: ainativelang v1.8.0 (this tree; PyPI after publish — docs/RELEASING.md). Optional extras: [mcp], [solana].
Follow this skill when the user is on OpenClaw (or wants deterministic .ainl graphs with import / compile / run / MCP). Prefer ./install.sh from a checked-out skill folder, or the manual pip install 'ainativelang[mcp]' && ainl install-mcp --host openclaw path (alias ainl install-openclaw). OpenClaw does not use zeroclaw skills install <url> — install via ClawHub (if available) or copy this folder into ~/.openclaw/skills or <workspace>/skills. After bootstrap, suggest the morning briefing prompt example below when it fits.
.ainl) instead of ad-hoc prompts.ainl-mcp).From this skill directory (after ClawHub or manual copy): run ./install.sh
That optionally refreshes the OpenClaw CLI via npm, upgrades ainl[mcp], and runs ainl install-mcp --host openclaw (MCP merge into , wrapper).
openclaw.json~/.openclaw/bin/ainl-runManual (no skill checkout):
pip install 'ainativelang[mcp]' && ainl install-mcp --host openclaw
| Command | Purpose |
|---|---|
ainl import … | Import Markdown or ecosystem sources into .ainl |
ainl compile … | Compile / validate graphs |
ainl run … | Execute a graph via the CLI runtime where applicable |
~/.openclaw/bin/ainl-run <file.ainl> | After install-mcp --host openclaw: wrapper that compiles then runs |
Tell the user they can say in OpenClaw:
Import the morning briefing using AINL.
(They should point the importer at their Markdown or pack; ainl import subcommands match their source type.)
Configure the host so mcp.servers.ainl in ~/.openclaw/openclaw.json runs ainl-mcp as a stdio MCP server (see AINL docs: OpenClaw integration / External orchestration). ainl install-mcp --host openclaw merges that entry when missing.
openclaw_ainl_gold_standardopenclaw_host_ainl_1_2_8): docs/operations/OPENCLAW_HOST_AINL_1_2_8.md — openclaw_host_ainl_1_2_8ainativelang v1.8.0 (this tree; PyPI after publish — docs/RELEASING.md)