Versioning & updates
Once an agent is published you keep it current by publishing new versions. Every new version goes through the same scan and review as a first submission — and your live version keeps serving the whole time. Rolling back is instant. The exact update path depends on the agent kind.
The one rule: your live version never goes down
Publishing a new version never takes your working agent offline. A new version is created as a separate pending version and reviewed on its own; the version buyers are already using keeps serving until the new one is approved. If a new version is rejected — or its build fails, for a code agent — nothing changes for your buyers. This is the same guarantee whether you edit a doer's tools, re-pull a code agent, or re-introspect an MCP-server listing.
A buyer's access is tied to their purchase, not to a specific version, so re-versioning and rolling back never revoke anyone's entitlement.
Which update path is yours
How you publish an update depends on the kind of agent. Each path produces a new pending version that goes to review; the current version keeps serving until it's approved.
| Agent kind | How you update it | What FindAgent re-runs |
|---|---|---|
| Doer / prompt recipe | Publish a new version — edit the system prompt, example prompts, and (for a doer) the whole tool surface, credential slots, and guardrails. | The security scan + the author-time credential-to-host binding check on any capability change. |
| Code agent · skills bundle | Re-pull from your connected GitHub repo (a “Publish new version → Re-pull from your repository” action, or the re-pull tool). | A fresh snapshot of your repo, the static scan, and — for a code agent — a rebuild in the sandbox. |
| MCP-server listing | Re-introspect the remote server so its listed tools match what your endpoint actually exposes now. | A sandbox-gated, SSRF-hardened re-listing of the remote server’s current tools. |
Doer & recipe agents — publish a new version
For a declarative doer or a prompt recipe you own, publishing a new version lets you edit the entire agent — not just the prompt:
- The
system_promptand the example prompts. - For a doer, the whole capability surface — add, remove, or edit tools and their
actionbindings, the credential slots, and the guardrails. You pass the complete intended tool list; it replaces the old one. - The price (once payments launch). A price change applies going forward — it is never retroactive, so existing buyers keep their access.
Any change to a tool action or credential re-runs the same security scan and the same author-time check that a credential slot's allowed_hosts cover the host its action reaches. A changelog is required (10–500 characters), though it's auto-drafted from the tool diff when you edit tools.
findagent_bump_version tool). Both go through the same review; see MCP & API.Code & skills agents — re-pull from GitHub
A code agent or a skills bundle is built from a snapshot of your repository taken at pull time, not its live state on GitHub. So editing GitHub alone doesn't change your published agent — you re-pull to bring your latest code into a fresh snapshot:
- Trigger a re-pull. From your dashboard use Publish new version → Re-pull from your repository(or the
findagent_repulltool). You can pull the latest commit on your stored branch, or a specific branch, tag, or commit. - FindAgent re-fetches and checks it. The repo is re-fetched with your connected GitHub token, re-snapshotted, and re-scanned; a code bundle is also rebuilt in the sandbox. The new snapshot is diffed against your live version.
- A changed source becomes a pending version. If the source changed, a new version is submitted for review with an auto-drafted changelog. If nothing changed, the re-pull is a no-op. Either way your live version keeps serving.
MCP-server listings — re-introspect
An MCP-server listing points at a remote server you run — FindAgent never runs it, it only lists what the server exposes. When your server's tools change, re-introspect the listing so the catalogue reflects them. FindAgent re-lists the remote server's current tools over a sandbox-gated, SSRF-hardened scan and, if the tool surface changed, submits a new version for review with an auto-drafted changelog. If the remote can't be reached or is auth-walled, nothing changes and you're told why. Only the tool listing updates — nothing about your server executes on FindAgent.
Rolling back
If a new version isn't what you wanted, you can roll a published agent of any kind back to a prior version. The live agent immediately serves that earlier version again. A rollback target must be a version that was published before — so a rollback never serves un-reviewed content — and it re-derives the connection facts (hosted vs. local, the agent kind, the category) from the version you roll to, so the listing stays consistent. Existing buyers keep their access.
Withdrawing a pending version
If a version is stuck in review and you want to submit a fresh one instead, you can withdraw the pending version. Withdrawing only ever touches a version that is still awaiting review — it never touches the live published version, which keeps serving.
What never changes across versions
- The slug. An agent's URL is permanent for SEO and existing links — it can't change after the first publish, across any number of versions.
- Existing entitlements. A purchase entitles the buyer to the agent, not to one version. Re-versioning, rollback, and price changes never revoke access someone already has.
- The review gate. Every new version — bump, re-pull, or re-introspect — passes the same automated scan and human review before it can go live. Nothing self-publishes.
Related
- Creator guide — the first-publish flow, the submit wizard, and the review pipeline.
- Code agents — build and readiness states, and re-pull for a code agent.
- Manifest spec — the tool, credential-slot, and guardrail shapes you edit in a new version.