Every InTouch Hub and ClawHub skill.
With the governance layer OpenClaw doesn't ship.
And skills are where you stop starting from a blank page. Find it and run it: install something that already works, point it at your setup, run it in your own language.
InTouch AI runs three tiers of skills: imports from upstream ClawHub (5,000+ OpenClaw skills, always converted to deterministic IML before execution), native MD skills (markdown that orchestrates the AI assistant's tools), and Workflow Files (Workflows as Code) (the code your senior engineers can actually read). All three tiers run under the same access control, the same encrypted credential vault (strong encryption, never exposed even to the AI), the same audit log, and the same alerting fabric. Discovery is OpenClaw. Execution is enterprise.
Find it. Build it. Share it.
The InTouch Hub is a library of 2,300+ ready-to-run tools — growing every week — plus workflows, skills, and monitors. Install one, point it at your setup, and run it in your own language. Don’t see what you need? Build it yourself, or have InTouch AI build it. Built something good? Hand it to your team in one step.
Find it
Browse the Hub for a ready-to-run tool, skill, workflow, or monitor. Install in seconds, point it at your credentials and data, and run — no blank page, no config. Each one converts to a native, deterministic InTouch artifact.
Build it
Two ways, no blank page. Author a new tool, skill, workflow, or monitor directly — or just describe what you want and the AI assistant generates, validates, and readies it for you. The thing you couldn’t find, built in plain language instead of code.
Share it
Export any tool, skill, workflow, or monitor as a folder or zip and hand it to a teammate or another InTouch install. They import it and run — it travels with its manifest and lands under their access control and credential vault. Governed sharing, not copy-paste. And because each IML artifact is a single self-contained file, you can skip the packaging entirely — send one .iml file (or a native SKILL.md) straight to another InTouch user who uploads it to their instance. A direct peer-to-peer universe that doesn’t depend on the Hub.
Three Skill Formats. One Runtime.
Different authors, different use cases, same deterministic execution path. This is the determinism dial: spin from pure AI to fully deterministic per skill, and graduate any one of them to zero-AI-cost, identical-every-time, fully-audited as it earns your trust. Nobody else gives you the dial.
OpenClaw Imports
Imported, never run raw. Browse or search ClawHub from inside InTouch AI (via the ClawHub view). Preview the SKILL.md. On install, the AI analyzes the skill once and generates a native IML tool or IML workflow. Subsequent runs are zero-LLM-cost, auditable, schedulable.
5,000+ skills, growing. Install from the ClawHub view in the InTouch UI.
MD Skills (Native)
A markdown file with YAML frontmatter, authored directly in InTouch AI. The AI assistant treats the markdown as instructions and orchestrates the 60+ platform tools to carry them out. Useful for skills that need the AI's flexibility on inputs but your tools for the work.
Lives inside your server. Versioned. access control-gated.
Workflow Files (Workflows as Code)
A full workflow in version-controllable IML. Tools, dependencies, output pipes, AI steps, schedules, alerts — all declared. Git-diffable. Code-reviewable. Runs deterministically without any AI in the critical path.
The form that survives "the author left the company."
OpenClaw Standalone vs InTouch AI
OpenClaw is a great skill discovery and AI-agent format. InTouch AI is the enterprise runtime it doesn't ship.
| Governance Layer | OpenClaw (standalone) | InTouch AI |
|---|---|---|
| Access control | None — user runs everything | access control — roles, areas, publisher permissions |
| Credentials | Shell env vars (export API_KEY=…) | Encrypted vault, never exposed to AI context |
| Audit trail | None | Full workflow log with timestamps, inputs, outputs |
| Scheduling | Manual invocation only | Seven native schedule types, file triggers, condition-driven Monitors |
| Alerting | None | Subscriber notifications across 8 channels |
| Concurrency | None | Collision detection, exclusive workflow locks |
| Ownership / multi-user | Single user on a local machine | Single-tenant, access control within the tenant¹ |
| Execution cost | LLM tokens on every run | Zero — converted workflows run deterministically |
¹ One server per organization; users within the tenant are access-controlled via access control. InTouch AI does not run multi-tenant — that's a design choice, not a gap. It's what makes the credential vault, audit trail, and access control model tractable.
Discover, Convert, Operate
OpenClaw skills are never run raw in InTouch AI. They are always converted to native InTouch AI automation first — an IML tool or an IML workflow, as the AI decides based on the skill's shape. ClawHub is a discovery and import source, not a runtime.
Discover
Browse or search ClawHub from inside InTouch AI (the ClawHub view). Preview SKILL.md before installing. Filter by highlighted / non-suspicious / version / recency.
Convert
The AI analyzes the skill's requires block, its execution pattern (REST, CLI, script), and its credentials. It generates an IML tool (if the skill is a single reusable operation) or an IML workflow (if it's a multi-step workflow). Incompatible patterns are flagged, not faked.
Operate
The converted artifact runs deterministically. Zero LLM cost per execution. Full logging, alerting, scheduling, access control. The AI was paid once; the workflow runs forever.
The Provider Is a Config Setting, Not an Architecture
The AI tool inside a converted skill — or inside any IML workflow, or anywhere the assistant runs — can target any of 9 providers: Anthropic, OpenAI, Google Gemini, Mistral, Groq, DeepSeek, xAI, Hugging Face, or Ollama. Swappable per tool. The workflow definition doesn't change; only the tool name and credential reference do.
If Anthropic refuses your domain (financial advice, legal guidance, medical synthesis) — swap to Ollama with a local open model like qwen2.5. Zero restrictions, zero API cost, no data leaving your network. Works air-gapped if you need it.
The skill doesn't need to be rewritten. The workflow doesn't need to be rebuilt. The anthropic tool is replaced with ollama, and the same IML runs on the same schedule.
Four Ways to Run a Skill
From the AI assistant
In the AI assistant chat (web UI, PWA, or Android app), type @ followed by the skill name. The skill runs and the result streams back into the conversation.
On a schedule
Attach a schedule directly to a skill. No workflow wrapper, no ceremony. Seven schedule types available.
From a trigger
A file or folder trigger (or a Monitor arm) invokes the skill with event data — new files, changed dates, trigger event type, or matched check output.
REST API
POST /api/skill/run with {name, input}. Integrate from anything that can HTTP: CI pipelines, portals, custom apps.
Why "Convert First, Run Later" Wins
AI is the architecture here, not a feature bolted onto an older core. The AI sits at the center; the vault, the schedules, the access control, the audit log all sit behind it. That's why the convert step is native and not a plugin — and why a competitor who retrofitted AI onto a static runtime can't reproduce it. You can't bolt on a center.
Deterministic
An AI that interprets a skill afresh every run can improvise. Converted IML doesn't. If you need a report to come out the same way every Monday, improvisation is exactly what you don't want.
Zero Per-Run AI Cost
OpenClaw standalone consumes LLM tokens on every invocation. A converted InTouch AI workflow consumes them zero times. Over a year of daily runs, that's the difference between a real line item and a rounding error.
Fails Predictably
When the IML workflow fails, the failure mode is a specific tool's exit code and error text. Compare to "the AI sometimes doesn't call the retry tool." Predictable failures are the ones you can fix.
5,000+ Skills. One Governance Layer.
Don't start from a blank page. Start with the free Personal edition, find a skill that already works, point it at your setup, and run it in your own language. Watch it convert to deterministic IML, run it on a schedule, and never pay an LLM bill for it again. This is AI automation — this is InTouch AI.