AI artifact sharing
Share AI artifacts. Turn feedback into revisions.
Give work from Claude, Codex and other AI tools a shared home. Send a link to an HTML prototype, Markdown report or slide deck, then bring the reviewer’s comments back to your agent.
Connect your agentOne workspace across AI tools
Connect your agents to the same tivq workspace. A report created with one tool can be read and revised by another, with a separate revocable key for each agent.
Feedback attached to the work
Reviewers can anchor feedback to HTML, Markdown, plain text, JSON or individual slides. Comments stay attached to the version reviewed, so your agent has the context for its next revision.
A link with a lifetime
Publish a temporary preview instead of maintaining another permanent site. Choose an expiry, share the link and revoke the resource when the review is finished. Anyone holding the link can access it until it expires or is revoked.
How it works
Connect Claude or Codex
Use the integration instructions to connect your tool to a workspace through MCP or the CLI.
Publish the artifact
Ask your agent to publish the HTML, Markdown, text or slides to tivq and return a share link.
Review and revise
Open the link, leave comments, then ask your agent to read the feedback and publish a new revision. Feedback does not wake the agent automatically.
Try this prompt
Publish this HTML prototype to tivq with a one-hour expiry. Return the share link. After I review it, read the feedback and publish a revised version.
Find connection steps in integrations and workflow details in the agent guide.
Questions
Can I share work made with Claude or Codex?
Yes. Your connected agent can publish supported HTML, Markdown, text and slide content to tivq. This creates a separate tivq resource; it does not change sharing settings inside your AI tool.
Do reviewers need a workspace key?
No. A recipient can open the share link in a browser without a workspace key. Treat that link as access to the resource: anyone you forward it to can use it.
Can my agent act on comments?
Yes. Ask the agent to retrieve feedback and create a revision. Reusable artifacts support comments and revisions; one-time resources, messages and instructions do not.