[PR #66] [MERGED] Update CLIP Text Encode node documentation #72

New issue

Closed

opened 2026-10-26 22:08:43 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-26 22:08:43 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/Comfy-Org/embedded-docs/pull/66
Author: @polydecay
Created: 12/1/2025
Status: ✅ Merged
Merged: 12/23/2025
Merged by: @comfyui-wiki

Base: main ← Head: pd-update-cliptextencode

📝 Commits (2)

10f9faa Add wildcards and comments to ClipTextEncode
53b1570 Refactor ClipTextEncode docs

📊 Changes

1 file changed (+54 additions, -36 deletions)

View changed files

📝 comfyui_embedded_docs/docs/ClipTextEncode/en.md (+54 -36)

📄 Description

This PR adds documentation for wildcards and comments to the CLIP Text Encode node as described in the main ComfyUI README here.

NOTE: I made two commits for this: the first is a minimal commit that adds wildcards and comments to the existing documentation. The second commit is a more opinionated refactor of the CLIP Text Encode documentation. Please let me know if you don't like the larger refactor, and I'll revert it to the minimal commit instead.

While adding wildcards and comments, I noticed the overall documentation could be improved, so I reviewed all of it. The refactor brings several improvements beyond adding the missing features; here are some of the main ones:

Reduced the overreliance on bullet points, which previously made the text read awkwardly.
Reduced overall "fluff" words (e.g., "like") to make the text more concise and professional.
Clarified that weights can be reduced by using values below 1.0.
Added instructions for including literal () in prompts.
Clarified that "prompt" corresponds to "description" in the input section for beginners unfamiliar with the terminology.
Removed general tips for writing prompts because best practices depend on the model (for example, some models work better with tag-based prompting).
Moved the "important" notes about embeddings into the embedding section to make them easier to find when skimming.
The EasyNegative embedding is supposed to be used in the negative prompt, so I made the surrounding words in the example negative as well.

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/Comfy-Org/embedded-docs/pull/66 **Author:** [@polydecay](https://github.com/polydecay) **Created:** 12/1/2025 **Status:** ✅ Merged **Merged:** 12/23/2025 **Merged by:** [@comfyui-wiki](https://github.com/comfyui-wiki) **Base:** `main` ← **Head:** `pd-update-cliptextencode` --- ### 📝 Commits (2) - [`10f9faa`](https://github.com/Comfy-Org/embedded-docs/commit/10f9faaab0c32d1d99f85b724bda8d7072a32171) Add wildcards and comments to ClipTextEncode - [`53b1570`](https://github.com/Comfy-Org/embedded-docs/commit/53b1570a9314e49d881e2f9b42b71560be59cb16) Refactor ClipTextEncode docs ### 📊 Changes **1 file changed** (+54 additions, -36 deletions) <details> <summary>View changed files</summary> 📝 `comfyui_embedded_docs/docs/ClipTextEncode/en.md` (+54 -36) </details> ### 📄 Description This PR adds documentation for wildcards and comments to the CLIP Text Encode node as described in the main ComfyUI README [here](https://github.com/comfyanonymous/ComfyUI?tab=readme-ov-file#notes). **NOTE**: I made two commits for this: the first is a minimal commit that adds wildcards and comments to the existing documentation. The second commit is a more opinionated refactor of the CLIP Text Encode documentation. Please let me know if you don't like the larger refactor, and I'll revert it to the minimal commit instead. --- While adding wildcards and comments, I noticed the overall documentation could be improved, so I reviewed all of it. The refactor brings several improvements beyond adding the missing features; here are some of the main ones: - Reduced the overreliance on bullet points, which previously made the text read awkwardly. - Reduced overall "fluff" words (e.g., "like") to make the text more concise and professional. - Clarified that weights can be reduced by using values below 1.0. - Added instructions for including literal `()` in prompts. - Clarified that "prompt" corresponds to "description" in the input section for beginners unfamiliar with the terminology. - Removed general tips for writing prompts because best practices depend on the model (for example, some models work better with tag-based prompting). - Moved the "important" notes about embeddings into the embedding section to make them easier to find when skimming. - The `EasyNegative` embedding is supposed to be used in the negative prompt, so I made the surrounding words in the example negative as well. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-26 22:08:43 +00:00