Back to browse
GitHub Repository
1 starsGo

Sharing LLM Artifacts with your team

by d_pang·Jul 3, 2026·2 points·0 comments

Post Description

A common challenge we found as our non technical team began to work with Claude Cowork was the ability to share HTML/markdown files with the rest of the team.

With LLMs building is easy but sharing artifacts is still difficult. Getting a prototype or report meant using a technical platform like Cloudflare or Vercel to host simple web pages. Setting up these pages was too complex for non-technical folks and a waste of engineering time. While Claude and ChatGPT have recently added the ability to share artifacts, they lack flexibility with access control and lock your files into their platform.

Dropway is a simple fix: drag your files into the dashboard (or deploy via CLI or MCP) and get a shareable URL in seconds. Access control is set per site: private email allowlist, org-only, or public. Every deploy is immutable and content-addressed, so versioning is automatic and rollback is one click.

We added CLI and MCP integration so Claude, Codex, and Cursor can deploy directly to a site. You can create, share, iterate without leaving your LLM of choice.

Source-available under FSL and self-hostable at https://github.com/danielpang/dropway. Or use our hosted option at https://dropway.dev

Happy to answer questions about the architecture or the MCP integration, and would love feedback from anyone who has run into this same friction.

Thank you in advance!

Similar Projects

SaaS●●Solid

A "content compiler" that turns LLM output into validated artifacts

Treating LLM output like compiler input — with typed style guides, required-section enforcement, and explicit Confidence/LostElements on transformations — is a clever, non-obvious approach that could actually raise the signal-to-noise on generated content. The product shows useful practical features (export to PDF/HTML/JSON, jurisdiction-aware legal drafting, slide generation), but the real test will be how maintainable and authorable those rule sets are in messy, real-world workflows.

Big BrainBold Bet
hardikparikh29
104mo ago