How agent-generated files are versioned, shared, hosted, and made interactive
When your Gumloop agents generate files during a conversation, those files are saved as artifacts. You can preview, download, share, and version reports, spreadsheets, images, code files, and HTML dashboards from the chat.
When an agent generates a file in its sandbox (using code execution, data processing, or any tool that produces a file), it exports the file using a built-in export tool. The export:
1
Saves the file
The file is written to secure cloud storage.
2
Creates a version
Gumloop records a versioned artifact for it.
3
Generates previews
Preview thumbnails and representations are built.
4
Shows it in chat
The file appears as a rich card in the conversation.
No export setup is required. Files appear in the conversation when the agent exports them.
When an agent exports a file, it appears as a card in the chat message showing its display name or filename, file type icon, and a thumbnail preview (for supported types). Click the card to open a side panel preview without leaving the chat.
Each artifact has its own shareable page at a unique URL. The viewer page shows:
Full-width file preview
Display name or filename, file type, and size
Who created it and which agent generated it
Links to view the source chat and agent (if you have access)
All files from a conversation are listed in the Files section of the chat sidebar. Click any file to preview it, open it in a new tab, copy its link, or download it.
On mobile, tapping a file card opens the viewer page in a new tab instead of a side panel, and auto-preview of new files is disabled so it doesn’t interrupt the chat.
Gumloop can render inline previews for many file types:
File Type
Preview
HTML (.html)
Interactive sandboxed preview
Images (.png, .jpg, .gif, .webp, .svg)
Native image viewer
PDF (.pdf)
PDF viewer
Spreadsheets (.csv, .xlsx, .xls)
Spreadsheet viewer
Presentations (.pptx, .ppt)
Slide viewer or PDF preview
Audio
Browser audio playback for supported media
Video
Browser video playback for supported media
Text & Code (.txt, .md, .py, .js, .json, etc.)
Syntax-highlighted text viewer
All other files
Download prompt with file size
Audio and video stream for playback, including files larger than 50 MB. Other files larger than 50 MB are download-only instead of being previewed inline. Playback depends on your browser’s support for the file’s media format and codec; download the file if it cannot play in the viewer.
Yes. Give the agent an artifact from the conversation or its link, then ask it to check or change the file’s name, hosted link, or sharing.
Example request
Result
”Rename this file to Quarterly report”
Changes the display name, not the downloaded filename
”Invite this person as a Viewer”
Shares the file if your permissions allow it
”Change this file’s hosted URL alias to quarterly-report”
Changes the alias on the available artifact host
Changes normally require your approval; checking the file does not. File permissions and organization restrictions still apply. Change thumbnails and password links in the artifact’s Settings.
Does Don't ask again apply only to this file?
No. Don’t ask again for this tool applies to artifact changes made by that agent, not just one file. Permissions still apply. There is currently no settings control to turn asking back on for Manage Artifact.
What if sharing fails after the agent renames the file?
The new name or link may already be saved even if sharing fails. Ask the agent to check the file’s current settings before retrying.
Owners and Editors can give an artifact a display name without renaming the underlying file. The name appears on artifact cards, in the viewer, and in search. Changing it does not change the downloaded filename, file contents, artifact link, or version history.
1
Open the artifact's Settings
Open the artifact, then select Settings.
The display name and thumbnail controls are at the top of Settings.
2
Edit the display name
Double-click the name, enter up to 256 characters, and press Enter or click outside the field to save. Press Escape to cancel. Clear the name to use the filename again.
Owners and Editors can upload a custom thumbnail for an artifact. One thumbnail applies to every version of that file until you replace it or return to the default preview. It changes the cover image, not the file’s contents, and does not create a new version.
1
Choose an image
In the artifact’s Settings, select Upload thumbnail and choose a PNG, JPEG, or WebP image. The upload limit is 5 MB.
2
Crop and save
Adjust the crop, then select Save thumbnail.
Crop an uploaded image before saving it as the file's thumbnail.
3
Return to the default preview when needed
Click the thumbnail in Settings, then select Use default preview to remove the custom image.
When an agent exports a file with the same filename multiple times in the same conversation, Gumloop automatically creates new versions instead of overwriting. This gives you a full history of how a file evolved during the conversation.
Each version shows:
Version number (v1, v2, v3, etc.)
Timestamp of when it was created
File size
Owners and Editors can open version history and select a version to view. The latest version is shown by default.
Agents are instructed to keep the same filename and let the system handle versioning. If you see files like report_v2.pdf instead of version 2 of report.pdf, you can update your agent’s instructions to tell it not to rename files for versioning.
Select Share from the artifact viewer or its options menu to open the Share artifact dialog. From top to bottom it has:
An Add people or teams field with a role picker inside it (Editor or Viewer) and an Add button.
Who can edit: the Owner and any Editors, who can manage the file.
Who can view: people and teams with Viewer access, then two separate rows. Organization is No access or Viewer. Public link access opens its own view, where Who can access public link is No public link, Anyone with the link, or Anyone with the password.
Copy share link, Add a password, and Add custom domain at the bottom. Once a password or custom domain is set, these read Edit password and Edit custom domain.
Settings shows the same editing and viewing lists, plus the Password protected link controls.
Viewers see a limited sharing list and can remove their own direct invitation, not someone else’s access or an Owner. Removing your invitation does not remove access you have another way, such as through your team.This also works for Viewers invited from another organization. A file invitation does not give them access to the source agent or conversation.
Organization and public link access grant viewing access only, even if an older organization or public grant previously allowed editing. They do not remove a person’s separate Owner or Editor access. To let someone edit a file, add them or their team as an Editor.
Setting
Who can view
No organization access and No public link
People with existing access, including direct invitations and access inherited from the source task or agent
A team added as a Viewer
Members of that team. Add the team as an Editor instead to let its members edit the file.
Organization set to Viewer
Members of the artifact’s organization
Public link access set to Anyone with the link
Anyone with the link, without a password or Gumloop account
Public link access set to Anyone with the password
Anyone with the link who enters the password
The organization and public link rows are independent, so you can turn on one without the other.
A Password protected link lets people outside the file’s existing audience view it with a password. Team or organization access can stay enabled at the same time: authorized members do not need the password, while other link visitors do.
1
Turn on the password link
Open Settings, then select Turn on beside Password protected link. You can also select Add a password in the Share artifact dialog.
2
Save a password
Enter a password of 6 to 256 characters and select Save. The password link stays off until a password is saved. Choosing Anyone with the password without one fails with “Set a password before you turn on the password link for this item.”
3
Share the link and password
Use Copy link in Settings or Copy share link in the Share dialog. Send the password separately using your usual secure channel.
Viewer
Needs the password?
The artifact’s owners or source task creator
No
Someone explicitly invited to the file, or invited by name to its source agent with valid file access
No
A team or organization member covered by the file’s own team or Organization access
No
Someone who only has broad team or organization access to the source agent, with no separate file access
Yes
Other link visitors, signed in or not
Yes
To give your whole team password-free access, add that team to the file’sWho can view list. Team membership or access to the source agent alone is not the same as sharing the file with the team.When a password is required, the file shows a Password field and an Unlock button. Unlocking grants viewing access only, not permission to edit the file.
If Public link access is Anyone with the link, saving a password narrows that access first: a team artifact switches to its team, and a personal artifact switches to invited people only. Choosing Anyone with the link later removes the password link. Turning the password link off removes password-based access and restores normal access inherited from the task or agent. It does not make the file public.
How long does an unlock last?
24 hours. After that, a viewer who still needs the password is asked to enter it again.
What happens when I change or turn off the password?
Changing the password invalidates existing unlocks. Select Turn off beside Password protected link to remove password-based access while keeping the file’s other access settings.
Can I keep a password link when I change the internal audience?
Yes. Switching between invited people, team access, and organization access keeps the password link. Choosing Anyone with the link removes it because that link no longer requires a password.
Can someone guess the password?
Repeated incorrect guesses are throttled. Correct unlocks are unaffected.
Can I see the saved password later?
Yes. People who can manage the artifact’s sharing can reveal the saved password in Settings.
Does password protection work on a custom domain?
Yes. A hosted URL points to the same artifact, so its viewing permissions and password requirement apply there too.
Your organization’s role settings can block external password-protected artifact sharing. If Turn on is blocked, ask an organization admin to review your role’s External password-protected artifact sharing permission.
Each agent has an Artifact Sharing setting that controls how new artifacts are shared when created. You can configure this in Access → Artifact Sharing → Default behavior:
Setting
What Happens
Default
Inherits access from the conversation and agent, subject to your organization’s conversation privacy policy. People who can read a conversation can access its files; team membership, Agent Owner access, or organization admin access alone does not expose files from a private conversation. See Agent access.
Organization
Files are shared with your entire organization
Anyone
Files are publicly accessible via link
The default sharing setting applies to new artifacts only. Owners and Editors can change an artifact’s sharing after it is created, subject to organization restrictions.
A public agent does not give signed-out visitors access to personal workspace files or files linked to a conversation. Only top-level team workspace files, with no conversation link, can inherit anonymous access from the agent’s public setting. A file’s own Anyone with the link or Password protected link setting is a separate way to share that file with outside visitors.
On Enterprise, editors and owners see an Analytics tab at the top of the artifact, next to Preview and Settings. It shows Views, Viewers, and Avg. time on page for the version and date range you pick (for example All versions and Last 30 days), with breakdowns by viewer, country or city, and device or browser. Visits by the person who created the artifact aren’t counted. Recent views lists each view, and you can filter it by viewer. Open a view to see the viewer and the tool calls the page ran, with status and latency. You can open a call’s inputs and outputs if you could see that call on the Connector activity page.
Each artifact has a canonical /artifacts/{id} URL. You can also give it a subdomain on gumloopartifacts.com:
your-artifact.gumloopartifacts.com
Use a custom domain to share a memorable link to a dashboard, report, or other artifact.
A custom domain is an alternate URL for the same artifact, not a separate copy or a published snapshot. Viewing permissions and version history still apply. Connector behavior differs between gumloopartifacts.com and organization-owned domains; see Your credentials, your data.
Hosting is configured per artifact. Open Settings, then select Turn on beside Custom domain. You can also select Add custom domain in the Share dialog. Once enabled, the custom URL appears in Settings and you can edit its alias.You need Owner or Editor access on the file to manage hosting. Hosting does not grant anyone additional viewing or editing access.When you enable hosting, Gumloop derives an alias from the artifact’s filename: the extension is dropped, anything that isn’t a letter or number becomes a hyphen, and the result is lowercased. If that alias is already taken, a suffix is added automatically so enabling never fails on a collision.
Filename
Derived alias
Q3 Pipeline Review.html
q3-pipeline-review
report.pdf
report
q3.html
q3-file
Very short names are padded to meet the three-character minimum, and long names are truncated.
You can rename the alias at any time while hosting is enabled. Aliases follow the same rules as hosted pages:
3–64 characters
Lowercase letters, numbers, and hyphens only
Must start and end with a letter or number
Must be unique across Gumloop
Certain aliases like admin, api, app, auth, beta, docs, gumstack, help, localhost, mcp, sandbox, staging, support, ws, and www are reserved and cannot be used.
Renaming frees the old alias immediately. Anyone holding the previous link gets a dead URL, and the old alias becomes available for someone else to claim.
Once Custom domain is on, Settings shows an Embed box under the custom URL. Select its copy button and paste the snippet into any web page, such as an internal wiki, a help center, or a company site:
The snippet lets the embedded page use the microphone, camera, clipboard, and full screen. Change the height to fit your page. Embedding doesn’t change who can open the artifact. If Public link access is No public link, the box notes “Embedded viewers sign in through a new tab,” and viewers sign in to Gumloop before the artifact loads.
Visitors sign in through Gumloop and return to the artifact, with two exceptions:
An artifact whose Public link access is Anyone with the link opens without sign-in, unless it has live-data scripts.
An artifact with live-data scripts always asks for sign-in, with any Gumloop account, because the scripts run as the viewer. If it’s password-protected, visitors need the password too.
No version pin. The custom domain always serves what /artifacts/{id} serves, so a new version of the artifact appears on both URLs at once.
No extra sharing. Hosting does not widen access. An artifact with No public link stays private on its custom domain. A public URL does not make the file public.
Canonical links still work. The /artifacts/{id} URL remains available. Settings offers the custom URL for copying once hosting is enabled; links to older versions remain version-specific.
Disabling hosting immediately stops the subdomain from resolving. The alias stays reserved for that artifact, so you can turn hosting back on later with the same URL. Deleting the artifact also stops the subdomain from resolving.
Enabling, renaming, and disabling artifact hosting are all recorded in audit logs.
Organization admins can claim one organization-level domain for artifact hosting from Organization Settings.
1
Claim your domain
In Organization Settings, open the Custom artifact domain section, select Add domain, and enter a domain such as yourdomain.com. New artifacts are shared on this domain once its DNS records are verified. The Add domain button stays disabled until the domain is valid.
2
Add DNS records
Add these records at your domain provider. DNS changes can take up to 48 hours.
Type
Host
Value
TXT
_gumloop-verify.<your-domain>
gumloop-verify=<token shown in the dialog>
CNAME
*.<your-domain>
cname.vercel-dns.com
3
Wait for verification
Verification runs against those records; use Re-check DNS to check again, or DNS records to see the values you need.
New artifacts are automatically shared on your own domain instead of gumloopartifacts.com.
You can paste the bare domain; casing, http:// or https:// prefixes, paths, and a leading *. are cleaned up for you. Gumloop-owned and other reserved suffixes are rejected, as is a domain another organization has already claimed. Each organization can have one domain at a time, so remove the current domain before adding a different one.
Newly exported artifacts from the organization’s agents are automatically given an alias at https://<alias>.<your-domain>, replacing the corresponding <alias>.gumloopartifacts.com URL.
The alias is derived and collision-suffixed using the same behavior documented for gumloopartifacts.com aliases.
Auto-hosting is best effort, so a hosting failure never fails the artifact export.
Auto-hosting is skipped when hosting was explicitly disabled for the artifact or when the artifact already has a hosted alias. An explicitly disabled artifact is not silently re-enabled.
When your organization has an active domain, enabling Custom domain in the artifact’s Settings hosts the file on that organization domain. The file’s sharing permissions still apply.An artifact with No public link stays private on the organization domain.
Paste an artifact link into chat and ask the agent to import it. It can use the original https://gumloop.com/artifacts/... link, a gumloopartifacts.com link, or a link on your organization’s artifact domain.When a specific version is named, it is used ahead of any version in the link. A link the agent can’t resolve or doesn’t have access to comes back as simply not found or not accessible, without revealing whether the artifact exists.
HTML files get special treatment. Agents can generate fully interactive HTML pages, dashboards, and web applications that render directly in the viewer.
HTML projects that use the agent’s default theme follow your Gumloop appearance in the viewer and your operating system’s appearance when opened standalone.Ask the agent to customize both light and dark colors. For direct edits, the palettes are in theme.css. Older files are not automatically restyled; ask the agent to update the file if needed.
Yes. An HTML artifact can embed another page from the same artifact bundle in an iframe. Ask the agent to create a multi-page HTML project with an embedded page, such as a report inside a dashboard. Embedded pages can use the artifact’s live-data scripts and follow its light or dark appearance. This does not allow embedding arbitrary external sites or pages from other artifacts.
HTML artifacts support a full screen mode that hides the toolbar and gives the artifact the full browser window. This is useful for dashboards, interactive tools, and presentations. Click the full screen button in the viewer toolbar to enter full screen.
HTML artifacts run in a strict security sandbox. This is important because agents can generate arbitrary HTML and JavaScript. The preview runs on an isolated origin, separate from your Gumloop session. It can read its own artifact files, but can’t read the Gumloop app’s cookies or local storage. A page can connect only to its own origin and to outside servers the agent declared when it saved that version.Gumloop serves each published page at the root of its own domain. Pages can use in-page routing with history.pushState, and unknown paths fall back to index.html. They can also use the microphone, camera, and clipboard after a click, embedding in other sites, and live-data scripts. They can open WebSockets and EventSource streams, and send fetch or XHR requests, to the servers the agent declared. The browser blocks connections to any other server. alert, confirm, and prompt dialogs work in the preview.Attached data-script calls use their declared tools, connection settings, and applicable tool policies.
Chat Widgets are compact charts, comparisons, or mockups inside a live web reply, using saved content or live data. Artifacts are exported files you can download, share, and version.
A Chat Widget appears inside the agent's reply. This order board uses fictional sample data.
Output
Where it appears
Can you share it separately?
Chat Widget
In the conversation, not in Files or your artifact library
No permanent standalone file link. Ask the agent to export a file if you want to reuse or share it separately.
Artifact
As a file card and in Files
Yes, using the file’s sharing and hosting settings
Enable Chat Widgets in the agent’s Agent tab under Abilities. It is on for newly created agents; existing agents need it enabled. Widgets appear only in live web chats, not Slack, Teams, email, or triggered tasks.An existing HTML artifact can also appear inline. That view keeps the artifact’s access and version settings; it does not create a new file or make the artifact public.
Yes. With Chat Widgets and Computer enabled, ask your agent for an inline chart or dashboard that loads data from your connectors. Live widgets are not available in incognito chats.Live widgets follow the same integration consent, connector account settings, and tool permissions as live artifacts. Opening a live widget reruns its data scripts and uses the viewer’s credits. Refreshes also use credits; data updates only when the widget loads or refreshes, not continuously in the background.
Interactive artifacts are HTML files with attached scripts that load live integration data. Live scripts require Computer and are not supported in incognito tasks. The connections those scripts can use depend on the source agent, the file’s sharing settings, and where you open it. See Your credentials, your data.
When you ask an agent to build something that needs live data, it creates two things:
An HTML file with the layout, styling, and JavaScript for the UI
One or more Python data scripts that fetch data from your integrations at view time
You don’t need to know how the two are wired together. Just ask your agent for a dashboard, report, or tool that uses your connected apps, and it handles the rest.
When you open an interactive artifact for the first time, you’ll see a consent overlay that lists every integration the file can access and the specific actions it can perform.
You must approve before any scripts run. This is a deliberate security step. It means:
You always know exactly which integrations a file uses
No data is fetched until you explicitly approve
If someone shares a file with you, you decide whether to grant it access to your accounts
If the page connects to outside servers, the overlay also lists them under It can also send data to these sites.
Your acknowledgement is saved for that artifact version. Review the listed integrations and actions before continuing.
An interactive artifact’s connections depend on where you open it and how the source agent’s connectors are configured. Sharing a file or giving it a password does not give visitors access to the creator’s personal connections.
Where you open the artifact
Which connections its scripts can use
Inside Gumloop or on gumloopartifacts.com
Your available connections, plus agent-owned connections configured on the source agent when the artifact is not public or password-shared
On an organization-owned artifact domain
Agent-owned connections configured on the source agent only, not your personal connections
For example, a dashboard using viewer-owned Slack connections can show each signed-in viewer data from their own Slack account. A connector configured as agent-owned uses the account pinned to the agent instead.
If the artifact has Anyone with the link access or a Password protected link, its scripts cannot use agent-owned connections. This also applies when a password link is kept alongside team or organization access. On an organization-owned artifact domain, that leaves no connector access for the scripts.
Your connector credentials are not exposed to the HTML itself. If a required connector is unavailable, complete the setup prompt before loading that data.
Anything that combines a UI with live integration data. Here are some example prompts:
What you want
Example prompts
Dashboards and monitors
”Build me a dashboard that shows my open Linear issues, today’s Google Calendar events, and my unread Gmail count." "Create a live team status page that pulls from Slack, Google Sheets, and HubSpot." "Make a monitoring page that shows my recent GitHub pull requests and their CI status.”
Interactive tools
”Create an HTML form where I can compose a message, pick a Slack channel from a dropdown, and send it." "Build a tool that lets me search my Google Drive files and preview them." "Make a form that creates a new Linear issue with title, description, and assignee fields.”
Reports with live data
”Generate a weekly summary report that pulls my Gmail activity, calendar meetings, and Slack messages from the past 7 days." "Create a CRM overview that shows my HubSpot deals pipeline with real-time data.”
Multi-integration pages
”Build a meeting prep page that, given a calendar event, pulls the attendee list from Google Calendar, finds their LinkedIn profiles, and shows recent email threads.”
The more specific you are about which integrations and data you want, the better the result. Tell the agent exactly which tools you want it to pull from.
A shared dashboard can show each viewer data from their own connections. For example, a “My Open Tasks” dashboard with viewer-owned connections shows each person their own tasks. Agent-owned connections instead use the account configured on the source agent, where that access is allowed.
Ask the agent to add a download button to an interactive artifact, for example to export the dashboard’s data as Excel, a filtered CSV, or a PDF. The button creates the file when you click it, and your browser downloads it.The script builds the file when you click the button and returns it as a browser download, rather than saving it as a new artifact.
Example prompt
What you get
”Add an Export to Excel button to this dashboard.”
A button that builds an .xlsx of the current numbers.
”Let me download the filtered rows as CSV.”
A button that exports whatever your filters selected.
A download runs a script, so it takes a few seconds and costs credits like any other data load. If the script produces nothing, the artifact shows an error instead of saving a file.
Creating the artifact uses the agent conversation’s usual credits. After export, live data scripts and AI model calls use credits billed to the viewer rather than the creator. Organization members use their organization’s credit balance and any applicable user limits. If a credit limit blocks execution, the script returns an error. See Credits.
Yes. A page’s data scripts can send the data they load to an AI model and show the answer on the page. The page can summarize, tag, score, or answer questions about live data every time someone opens it, without anyone starting a chat.Example: a support triage board. Ask an agent:
“Build a dashboard of this week’s open Zendesk tickets. Tag each ticket’s sentiment as positive, neutral, or negative, and its product area. Put a three-sentence summary of the main themes at the top, and add a box where I can ask questions about these tickets.”
When a teammate opens the page:
A data script loads the open tickets with that teammate’s Zendesk connection.
A model tags each ticket and writes the summary. Each tag and the summary appear next to the data.
Typing a question into the box, such as “Which customers mentioned billing twice?”, sends the question and the loaded tickets to a model and shows the answer on the page.
Everyone who opens the page gets fresh results from their own data, with no chat or prompt to write.More ideas:
What you want
Example prompt
Summaries
”Add a two-line summary under each deal in this pipeline dashboard, using its latest emails and notes.”
Tags and verdicts
”Flag every invoice in this sheet that looks like a duplicate or is missing a PO number.”
Scores
”Score each inbound lead from 1 to 5 for fit with our ideal customer profile and sort by score.”
Questions about the data
”Add an Ask box so viewers can ask questions about this month’s numbers.”
Drafts
”Next to each overdue task, draft a short follow-up message the viewer can copy.”
Pages can use two kinds of models. Chat models write free text, such as summaries, answers, and drafts. Decision models answer closed questions like yes or no, pick one option, or a score, and suit tagging many rows quickly.Who can use it: the person who publishes the page needs a custom role with AI models from agent code turned on. The feature is off by default. Every run also checks the viewer’s own role and model access, and bills the model usage to the viewer. There is no per-run approval. If an older page can’t call models, republish it.
Opening an artifact again can rerun its data scripts. A refresh button or auto-refresh timer can also start another execution.Each script execution uses credits. Long-running queries can time out; ask the agent to reduce the data scope if needed.
If a data script fails, the HTML receives an error response. Common causes are a disconnected integration, an API error, or a script timeout. Well-built artifacts show a friendly error message. If the agent didn’t include error handling, the section may simply be blank.Common causes and fixes:
Open Artifacts in the sidebar to find files across your agent conversations. Your personal library is available at Artifacts.
Tab
What it shows
Mine
Artifacts you created in your own agent conversations
Shared with me
Artifacts shared with you directly, through your teams, or through your organization
Search by display name or filename, filter by file type, and sort by date. A team artifact list also supports filtering by Creator. Each card shows the display name or filename, thumbnail, file type, and version.
By default, files created during an agent conversation are scoped to that conversation. However, agents can also work with workspace files that persist across conversations.Files saved to the /home/user/.workspace/ directory in the agent’s sandbox are treated as workspace-scoped artifacts. These files persist across conversations with the same agent.
Shared agent workspace, /home/user/.workspace/agent/: Access follows your role on the agent. Owners can edit shared files; Users can read them. Team membership is not an additional requirement for this shared disk.
Personal workspace, /home/user/.workspace/personal/: Private to you and persistent across your conversations with the same agent. It remains writable when the shared workspace is read-only.
This is useful for agents that maintain ongoing project files, configuration, or reference data that should carry over between sessions.
Sharing an agent or task does not automatically share your personal workspace.
Artifacts are stored in secure cloud storage. You can open them in the Gumloop viewer or download them. Workspace files remain available across conversations with the same agent.
Can I delete an artifact?
Yes. Anyone with Owner or Editor access to the file can delete it. Deleting is a soft delete: the file is removed from your files list, search, sharing, and normal viewer/API access, while its stored versions are preserved.
Can I share a file with someone outside my organization?
Yes. Invite them by email, choose Anyone with the link, or enable a Password protected link for other link visitors. Your organization can restrict external sharing. Organization, public link, and password access grant viewing access only.
Why can't I see the 'View chat' or 'View agent' link on a shared file?
These links only appear if you have access to the source agent. Artifact access does not automatically grant access to the agent that created it. Ask the agent owner to share the agent with you separately.
What's the difference between a static artifact and an interactive artifact?
A static artifact shows saved content, such as a PDF, image, or CSV. An interactive artifact can use attached scripts to load live data. An artifact you already acknowledged may open without a consent overlay.
Do interactive artifacts use my credentials or the creator's?
Inside Gumloop and on gumloopartifacts.com, scripts can use your available connections and eligible agent-owned connections. Organization-owned artifact domains use agent-owned connections only. Public or password-shared artifacts cannot use agent-owned connections. See Your credentials, your data for the full comparison.
Can an interactive artifact perform actions (send messages, create issues) or only read data?
Both. Agents can build forms and buttons that send Slack messages or create Linear issues. These actions use the connections described in Your credentials, your data and run a data script, just like read operations.
Can the HTML in an interactive artifact access my browsing session or cookies?
HTML artifacts cannot read the Gumloop app’s session, cookies, or local storage. They run on an isolated origin and can read their own artifact files. External data access uses controlled script or proxy paths, not the app’s browsing session.
My interactive artifact shows an error or blank section. What should I try?
Check that the required connectors are available. Refresh an expired connector from Connectors, check your credit limit, or ask the agent to reduce the data scope for a query that times out.
What happens before DNS verifies my organization domain?
Your artifacts keep working at their canonical URLs and their gumloopartifacts.com URLs. Nothing is served on the new organization domain until its status is Active.
Can my organization claim more than one artifact domain?
No. An organization can claim one custom artifact domain at a time. Remove the current domain before adding a different one.
Does a custom domain make private artifacts public?
No. A custom domain is an alternate URL for the same artifact. An artifact with No public link stays private and uses the same access checks as its canonical URL.
Can an agent import an artifact from a link?
Yes. Paste an artifact link into a chat and the agent can pull that file into its sandbox.
What happens if the organization domain is removed?
Links on that domain stop working. Canonical artifact URLs continue to work.
