Lmcq Auth Workflow Encryption

A ComfyUI workflow is just a JSON file - anyone who has it can open it, read every node and setting, and copy your entire pipeline. If you've built something you want to sell or gate rather than give away for free, this node encrypts that JSON behind a password so it can only be opened by someone running LmcqAuthWorkflowDecryption with the right key.

Why this matters more than it sounds like. ComfyUI's culture leans hard toward "workflow included" - sharing the actual graph, not just the output image, is the norm, and posts that withhold workflows tend to draw pushback in the community. That's great for open collaboration and genuinely bad if you've spent weeks tuning a complex multi-stage pipeline you'd like to charge for. This node is the pack's answer to that tension: keep the convenience of a shareable ComfyUI graph while still gating who can actually open it.

The fields. workflow_file is a dropdown - pick the workflow you want to protect from either your ComfyUI root workflows folder or the plugin's own workflows folder (per the README, both show up, labeled root/filename or plugin/filename). key is the password; whoever decrypts needs to enter it exactly. save_name (default encrypted_workflow) names the output. machine_codes is an optional multiline field - leave it blank for password-only protection, or paste in hardware codes (gathered via LmcqGetMachineCode from each authorized user) to also restrict decryption to specific machines, one code per line. This is a terminal, output-only node - no wired outputs, it writes the encrypted file next to the original.

What "Auth" means for this node specifically. This sits in LMCQ's cloud-backed authentication system rather than its purely-local "Runtime" family. Per the README, that system is built around a hosted platform where creators register with an auth_key/auth_secret, and decryption on the recipient's end can involve real-time validation against that server - usage limits, expiration windows, revocation - rather than a static local password check. That's more capable if you're managing an ongoing sale relationship with many buyers, but it also means the person decrypting your workflow may need network access to LMCQ's platform for it to work at all, which the pack's simpler LmcqRuntimeWorkflowEncryption variant (not covered here) doesn't require.

Installing it. Through ComfyUI Manager (search "ComfyUI-LMCQ") or:

cd ComfyUI/custom_nodes
git clone https://github.com/sebord/ComfyUI-LMCQ

Restart ComfyUI afterward.

Worth knowing before you build a business on this. Custom nodes run with full, unsandboxed access to whatever machine they're installed on, and this pack's core protection logic ships as deliberately obfuscated, compiled code by design - the README frames this explicitly as hardening against reverse-engineering. That's the entire point of an encryption node, but it does mean neither you nor your buyers can actually read what the code is doing, which is a bigger trust ask than most custom nodes. Nothing about this pack's track record suggests bad behavior, but if you're about to build a paid product around it, it's worth knowing you're trusting a small team's closed implementation, not an auditable one.

Troubleshooting. The recurring, reported problem with this entire pack isn't encryption logic misbehaving - it's the pack failing to load at all. A compiled runtime module the Auth nodes depend on has gone missing on some installs, throwing a ModuleNotFoundError (runtime.model_protection or api_model_protection) at startup, and because that import lives in the pack's shared __init__.py, it takes down every Lmcq node at once, not just this one. If LmcqAuthWorkflowEncryption isn't showing up in your node search, check your console log for that error before assuming anything else is wrong. The pack's January 2026 changelog documents the fix: copy the Python-version-matching runtime files into the runtime folder, remove the mismatched ones, restart.