We're closing the loop. Versioning, experimental mode, and share links now cover commands and workflows—the last two of SuperClawd's four entity types. If you've used these on skills or agents, there is nothing new to learn: it's the same model, built on the same shared core, now reaching everything you build.
The Same Proven Model, Now Everywhere
When we shipped versioning for skills, the goal was simple: bring the safety and control you expect from shipping code to the instructions your team runs in Claude Code. Then we extended it to agents on a shared internal core. Today that core reaches the rest of the suite—commands and workflows—so all four entity types behave identically.
| Git | SuperClawd versioning |
|---|---|
| Working directory | Draft — the live, always-editable entity |
| A commit | Release — a frozen, immutable snapshot |
git commit | New Release — freeze the draft into a release |
A stable tag | The stable pointer — the one release everyone runs |
git restore | Restore to draft — bring an old release back to edit |
The first release you ever cut auto-promotes to stable, so an entity becomes servable the moment it has its first release. Promote points the stable tag at any release—newest is a normal ship, older is an instant rollback. Restore to draft copies a chosen release back into your live draft so you can evolve a known-good version forward. And the safety guarantee that makes all of this trustworthy still holds: an unreleased entity is never served. If a command or workflow has no promoted stable release, Claude Code skips it—it does not fall back to your draft.
What's worth dwelling on is what a "snapshot" actually captures for these two new types.
Command Releases: Body and Arguments, Snapshotted Together
A command is more than its body—it's the body plus the arguments it accepts. A command release snapshots both together, as one atomic unit.
That matters because of a subtle consistency trap. A command shows up in two places: the slash-command hint in the catalog (the name and arguments Claude Code surfaces) and the body that actually runs when you invoke it. If those two could drift apart—if the hint advertised one set of arguments while the body expected another—you'd get confusing, broken invocations.
So they can't drift. Both the catalog hint and the executed body honor the same release/experimental gate. Whatever stable release is being served, the advertised arguments and the body that runs always come from the same snapshot. The hint always matches the body. Cut a release, and the version your team sees in the catalog is exactly the version that executes.
Workflow Releases: The Whole Graph
A workflow is a graph—nodes and the edges between them. A workflow release snapshots the entire graph: every node and every edge, frozen together into one immutable release.
Everything else is identical to skills and agents. Promote a stable release for the whole team. Roll back by promoting an older one—the full graph reverts instantly. Restore a release to draft to keep building on a known-good version. Your draft graph can be a half-rewired work in progress; it doesn't matter, because the stable release is what gets served, and an unreleased workflow is never served at all.
One Experimental Mode, Every Entity
Sometimes you do want to run a draft command or workflow against real Claude Code sessions before releasing it. That's what Experimental mode is for, and it's protected by the same two-key gate you already use everywhere else—both keys must be turned:
- Flag the entity experimental — a per-entity toggle on that specific command or workflow.
- Turn on Experimental mode in the CLI — the single global "Experimental mode" toggle in
superclawd config, off by default. This is the same toggle for every entity type. One switch for the whole machine gates experimental skills, agents, commands, and workflows alike.
Only when both are on does that entity serve its live draft—and only on that one machine. Every other machine, and every other entity, keeps running the frozen stable release. So one developer can dogfood one risky workflow against real sessions while the rest of the team stays on stable.
When the draft is good: cut a New Release, Promote it, and unflag experimental.
Sharing: One Link, Another Workspace
A great command or workflow you built shouldn't be trapped in your workspace. Share links let you hand one out via a single unlisted, revocable URL—the same sharing model skills and agents already have.
A share link hands out a released snapshot, never your live draft. When you create one, you choose how it tracks your releases:
| Mode | Recipients get | Behavior |
|---|---|---|
| Pinned | a specific frozen release | Frozen to whatever was stable when you made the link. Your later releases don't change it. |
| Live | your current stable | New visitors always get your latest stable release. |
You can optionally set an expiry date and a max imports cap, and you can revoke any link at any time to kill it immediately.
The Recipient Flow
The recipient page is now shared across all four entity types—skills, agents, commands, and workflows get the same safe, preview-first experience at /s/<token>:
- The recipient opens your link and sees a preview of exactly what they'd import.
- To import, they sign in and pick a target workspace.
- SuperClawd creates a brand-new, independent copy in their workspace—fully their own, with no link back to your workspace data.
Example: A Safe Iteration Loop
- Edit your workflow's draft graph freely all week.
- Flag it experimental and turn on Experimental mode on your machine to dogfood the draft live.
- Happy with it? Cut Release v5 with notes describing what changed—the whole graph freezes into the snapshot.
- Promote v5 to stable—now the whole team runs it.
- A bug slips through? Promote v4 to roll back instantly while you fix the draft.
- Want to share it outside your team? Create a share link, pinned or live, and send the URL.
History intact, team safe, no guesswork. Same for a command—where the body and its arguments travel together so the hint always matches what runs.
Get Started
Command and workflow versioning and sharing are available now in all SuperClawd workspaces.
- Go to app.superclawd.com
- Open a command or workflow and click Versions
- Cut your first Release and watch it auto-promote to stable
Want the full mechanics—the stable pointer, promote/rollback, restore-to-draft, the two-key experimental gate, and share-link modes? Read the docs at docs.superclawd.com.
Pro tip: This completes the set. The same release model now runs across skills, agents, commands, and workflows on one shared core—so the muscle memory you built on skills works everywhere. Cut a release before each workday's edits and future-you will love having a clean point to roll back to.