In8 template batch2 and new workflow templates#1006
Open
comfyui-wiki wants to merge 18 commits into
Open
Conversation
Updated via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Created via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Created via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Created via ComfyUI Template Manager
Updated via ComfyUI Template Manager
Move workflow and thumbnail back to templates/, re-add to media-video bundle, and sync manifest after unarchiving Wan Dancer (cloud-only).
Contributor
ComfyUI Node Compatibility ReportInformational only — this check does not block merging. Baseline: static scan of ComfyUI source — reports removed/deprecated core nodes only. Status: review suggested — some templates may use removed, deprecated, or outdated node settings.
Baseline: mode Scan warnings
At a glance
Issue types
What each issue type means
Details🟡 Warnings — 13 template(s), 20 finding(s). Deprecated nodes — migrate when convenientThis comment is automatically updated by the ComfyUI Node Compatibility Report workflow. |
Contributor
🔤 Spellcheck Results
❌ Workflow notes errors
|
Contributor
❌ Broken Links Found
Excluded: ✅ Whitelist Skipped URLs (6 URLs)These URLs are excluded from checking via whitelist.json skip_urls:
👻 Excluded Links (0 links)These links are intentionally excluded from checking and may require authentication or have known issues: Fix: Update or remove broken URLs in workflow files. Check workflow artifacts for full report. |
Contributor
❌ Template Validation Failed |
Contributor
|
✅ Upload JSON updated - 608 assets |
Contributor
|
❌ Upload JSON generation failed |
Contributor
❌ Template Validation Failed |
ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Pro Plus Run ID: ⛔ Files ignored due to path filters (5)
📒 Files selected for processing (46)
💤 Files with no reviewable changes (13)
📝 WalkthroughWalkthroughThe PR refreshes localized and archived template catalogs, updates several ComfyUI workflow graphs, adds three INT8 image workflows with preview assets, registers changed assets and checksums, and adds two PNG input records. ChangesTemplate catalog and localization
Possibly related PRs
Suggested reviewers: 🚥 Pre-merge checks | ✅ 2✅ Passed checks (2 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
✨ Simplify code
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 19
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
templates/index.fa.json (1)
1405-1411: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winThumbnail should point to the preview image, not the workflow JSON.
api_kling_o3_imagealready hasapi_kling_o3_image-1.webp, sothumbnailshouldn’t stay onoutput/api_kling_o3_image.json. Wire it to the bundled image asset instead, or the gallery will wear a paper bag.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.fa.json` around lines 1405 - 1411, Update the api_kling_o3_image entry’s thumbnail field to reference the bundled preview asset api_kling_o3_image-1.webp instead of output/api_kling_o3_image.json, while leaving the surrounding workflow metadata unchanged.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@bundles.json`:
- Around line 574-579: Add the existing video_wan_dancer entry from bundles.json
to packages/core/src/comfyui_workflow_templates_core/manifest.json, preserving
the manifest’s established structure and ordering so
test_manifest_aligns_with_bundles_config passes and the asset resolves at
runtime.
In `@scripts/data/i18n.json`:
- Around line 793-796: Remove the video_wan_dancer entry from the
_status.pending_templates inventory in scripts/data/i18n.json, since its title
and description are already localized for all required languages. Do not alter
the completed translation entries.
In `@templates/image_boogu_image_0_1_edit.json`:
- Around line 347-369: Update the MarkdownNote node’s widgets_values text in
“Faster Alternative” so the Boogu-Image-0.1-Edit-Turbo model name has balanced
Markdown bold markers, preferably wrapping the complete name with ** on both
sides. Leave the surrounding help note content unchanged.
In `@templates/image_qwen_image_edit_2509.json`:
- Around line 129-133: Remove the orphaned lora_name input from the relevant
subgraph node in image_qwen_image_edit_2509.json, then reindex subsequent input
ports so the node no longer exposes the unused slot. Preserve the functional
lora_name_1 input and its link 742.
In `@templates/image_qwen_image_edit_2511_int8.json`:
- Line 1836: Correct the user-visible group title in the template by changing “4
Steps Ligtning LoRA” to “4 Steps Lightning LoRA”.
- Around line 438-448: Rename the exposed BOOLEAN input label from
enable_turbo_mode to enable_4steps_lora, keeping its link ID 385 and existing
behavior unchanged.
In `@templates/image_qwen_image_edit_2511.json`:
- Line 1276: Update the “Note: Model Links” block to include
qwen_image_edit_2511_fp8mixed.safetensors alongside the existing model entry,
keeping the default UNET filename in the loader aligned with the documented
download.
In `@templates/image_z_image.json`:
- Line 341: Update the subgraph name in the relevant template metadata from
“Text to Image(Z-Image-Base Int8)” to identify the bf16 base graph, removing the
“Int8” designation while leaving the UNETLoader configuration and separate INT8
template unchanged.
In `@templates/index.ar.json`:
- Line 14459: Remove the duplicate "توليد النص" value from the tags entry in the
source template, leaving a single tag, then regenerate the mirrored
templates/index.ar.json and templates/index.json entries so both outputs stay
consistent.
In `@templates/index.es.json`:
- Around line 241-242: Translate the title and description for the Qwen Image
Edit 2511 Int8 catalog entry into Spanish, matching the terminology and style of
nearby localized entries such as “Qwen Image Edit 2511 - Reemplazo de
Materiales.” Preserve the model name, quantization details, input/output
behavior, use cases, and JSON structure.
- Line 8931: Localize the English title and description for
template_seedance2_0_viral_videos_character_swap and the description for
template_qwen_image_edit_2511_systms_action within the new “Casos de Uso”
section of templates/index.es.json. Preserve each entry’s meaning while
translating the user-facing text into Spanish.
- Around line 11-12: Translate the English title and description for the newly
added Z-Image Int8 and Boogu entries in the Spanish catalog into natural
Spanish, matching neighboring localized terminology. Update all affected catalog
references consistently across workflows and localized files while preserving
the existing metadata and structure.
- Line 1516: Update the thumbnail entry in the image comparison configuration to
use input/bold_outfit_woman.jpeg as the first frame, while retaining
output/image_flux2_klein_image_edit_9b_distilled.png as the second frame. Do not
use input/handbag_white.png, since it is only a reference image.
In `@templates/index.fa.json`:
- Line 1516: Update the thumbnail entry for the compareSlider variant in
templates/index.fa.json to reference the distinct -1.webp and -2.webp assets in
the two array positions, replacing the duplicated image path while preserving
the existing thumbnail structure.
In `@templates/index.pt-BR.json`:
- Around line 9-34: Localize the newly added INT8 template titles and
descriptions in the Portuguese catalog, including image_z_image_int8,
image_boogu_image_0_1_edit_int8, and image_qwen_image_edit_2511_int8, plus the
identified new “Casos de Uso” entries. Replace their English user-facing strings
with Portuguese translations while preserving all metadata and non-text fields.
- Line 15743: Correct the title value in the relevant template entry from “Bira:
Remover Fundo da Imagem” to “Bria: Remover Fundo da Imagem”, keeping the
existing capitalization and surrounding translation unchanged.
- Line 14459: Remove the duplicated "Geração de Texto" value from the tags array
for this multimodal Gemma4 entry, replacing it with the intended distinct
image/vision category if that category is defined elsewhere in the template;
otherwise retain only one text-generation tag.
- Line 1516: Update the thumbnail array for the image edit entry in
templates/index.pt-BR.json by replacing its first duplicated
output/image_flux2_klein_image_edit_9b_distilled.png value with the
corresponding input asset path, while keeping the second output thumbnail
unchanged so the compare view displays before versus after.
In `@templates/index.ru.json`:
- Around line 11-12: Translate the title and description fields for the three
INT8 entries image_z_image_int8, image_boogu_image_0_1_edit_int8, and
image_qwen_image_edit_2511_int8 in the Russian catalog. Match the Russian
terminology and style used by their corresponding non-INT8 sibling entries,
while preserving the existing model and workflow details.
---
Outside diff comments:
In `@templates/index.fa.json`:
- Around line 1405-1411: Update the api_kling_o3_image entry’s thumbnail field
to reference the bundled preview asset api_kling_o3_image-1.webp instead of
output/api_kling_o3_image.json, while leaving the surrounding workflow metadata
unchanged.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 1a3e4d9f-3618-46d0-87c1-1e48d6fed057
⛔ Files ignored due to path filters (5)
input/giant_tulip_wanderer.pngis excluded by!**/*.pnginput/lily_girl_blue.pngis excluded by!**/*.pngoutput/image_boogu_image_0_1_edit_int8.pngis excluded by!**/*.pngoutput/image_qwen_image_edit_2511_int8.pngis excluded by!**/*.pngoutput/image_z_image_int8.pngis excluded by!**/*.png
📒 Files selected for processing (46)
archived/archived_i18n.jsonarchived/index.ar.jsonarchived/index.es.jsonarchived/index.fa.jsonarchived/index.fr.jsonarchived/index.ja.jsonarchived/index.jsonarchived/index.ko.jsonarchived/index.pt-BR.jsonarchived/index.ru.jsonarchived/index.tr.jsonarchived/index.zh-TW.jsonarchived/index.zh.jsonbundles.jsonpackages/core/src/comfyui_workflow_templates_core/manifest.jsonscripts/data/i18n.jsontemplates/api_bytedance_seedream_5_0_pro_image_edit.jsontemplates/api_bytedance_seedream_5_0_pro_t2i.jsontemplates/image_boogu_image_0_1_edit.jsontemplates/image_boogu_image_0_1_edit_int8-1.webptemplates/image_boogu_image_0_1_edit_int8-2.webptemplates/image_boogu_image_0_1_edit_int8.jsontemplates/image_qwen_image_edit_2509.jsontemplates/image_qwen_image_edit_2511.jsontemplates/image_qwen_image_edit_2511_int8-1.webptemplates/image_qwen_image_edit_2511_int8-2.webptemplates/image_qwen_image_edit_2511_int8.jsontemplates/image_z_image.jsontemplates/image_z_image_int8-1.webptemplates/image_z_image_int8.jsontemplates/index.ar.jsontemplates/index.es.jsontemplates/index.fa.jsontemplates/index.fr.jsontemplates/index.ja.jsontemplates/index.jsontemplates/index.ko.jsontemplates/index.mcp.jsontemplates/index.pt-BR.jsontemplates/index.ru.jsontemplates/index.tr.jsontemplates/index.zh-TW.jsontemplates/index.zh.jsontemplates/video_wan_dancer-1.webptemplates/video_wan_dancer.jsonworkflow_template_input_files.json
💤 Files with no reviewable changes (13)
- archived/index.zh-TW.json
- archived/index.zh.json
- archived/index.ja.json
- archived/index.fa.json
- archived/index.json
- archived/index.ar.json
- archived/index.tr.json
- archived/index.pt-BR.json
- archived/index.es.json
- archived/index.ru.json
- archived/index.ko.json
- archived/index.fr.json
- archived/archived_i18n.json
There was a problem hiding this comment.
Caution
Inline review comments failed to post. This is likely due to GitHub's internal server error or limits when posting large numbers of comments. If you are seeing this consistently it is likely a permissions issue. Please check "Moderation" -> "Code review limits" under your organization settings.
Actionable comments posted: 19
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
templates/index.fa.json (1)
1405-1411: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winThumbnail should point to the preview image, not the workflow JSON.
api_kling_o3_imagealready hasapi_kling_o3_image-1.webp, sothumbnailshouldn’t stay onoutput/api_kling_o3_image.json. Wire it to the bundled image asset instead, or the gallery will wear a paper bag.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.fa.json` around lines 1405 - 1411, Update the api_kling_o3_image entry’s thumbnail field to reference the bundled preview asset api_kling_o3_image-1.webp instead of output/api_kling_o3_image.json, while leaving the surrounding workflow metadata unchanged.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@bundles.json`:
- Around line 574-579: Add the existing video_wan_dancer entry from bundles.json
to packages/core/src/comfyui_workflow_templates_core/manifest.json, preserving
the manifest’s established structure and ordering so
test_manifest_aligns_with_bundles_config passes and the asset resolves at
runtime.
In `@scripts/data/i18n.json`:
- Around line 793-796: Remove the video_wan_dancer entry from the
_status.pending_templates inventory in scripts/data/i18n.json, since its title
and description are already localized for all required languages. Do not alter
the completed translation entries.
In `@templates/image_boogu_image_0_1_edit.json`:
- Around line 347-369: Update the MarkdownNote node’s widgets_values text in
“Faster Alternative” so the Boogu-Image-0.1-Edit-Turbo model name has balanced
Markdown bold markers, preferably wrapping the complete name with ** on both
sides. Leave the surrounding help note content unchanged.
In `@templates/image_qwen_image_edit_2509.json`:
- Around line 129-133: Remove the orphaned lora_name input from the relevant
subgraph node in image_qwen_image_edit_2509.json, then reindex subsequent input
ports so the node no longer exposes the unused slot. Preserve the functional
lora_name_1 input and its link 742.
In `@templates/image_qwen_image_edit_2511_int8.json`:
- Line 1836: Correct the user-visible group title in the template by changing “4
Steps Ligtning LoRA” to “4 Steps Lightning LoRA”.
- Around line 438-448: Rename the exposed BOOLEAN input label from
enable_turbo_mode to enable_4steps_lora, keeping its link ID 385 and existing
behavior unchanged.
In `@templates/image_qwen_image_edit_2511.json`:
- Line 1276: Update the “Note: Model Links” block to include
qwen_image_edit_2511_fp8mixed.safetensors alongside the existing model entry,
keeping the default UNET filename in the loader aligned with the documented
download.
In `@templates/image_z_image.json`:
- Line 341: Update the subgraph name in the relevant template metadata from
“Text to Image(Z-Image-Base Int8)” to identify the bf16 base graph, removing the
“Int8” designation while leaving the UNETLoader configuration and separate INT8
template unchanged.
In `@templates/index.ar.json`:
- Line 14459: Remove the duplicate "توليد النص" value from the tags entry in the
source template, leaving a single tag, then regenerate the mirrored
templates/index.ar.json and templates/index.json entries so both outputs stay
consistent.
In `@templates/index.es.json`:
- Around line 241-242: Translate the title and description for the Qwen Image
Edit 2511 Int8 catalog entry into Spanish, matching the terminology and style of
nearby localized entries such as “Qwen Image Edit 2511 - Reemplazo de
Materiales.” Preserve the model name, quantization details, input/output
behavior, use cases, and JSON structure.
- Line 8931: Localize the English title and description for
template_seedance2_0_viral_videos_character_swap and the description for
template_qwen_image_edit_2511_systms_action within the new “Casos de Uso”
section of templates/index.es.json. Preserve each entry’s meaning while
translating the user-facing text into Spanish.
- Around line 11-12: Translate the English title and description for the newly
added Z-Image Int8 and Boogu entries in the Spanish catalog into natural
Spanish, matching neighboring localized terminology. Update all affected catalog
references consistently across workflows and localized files while preserving
the existing metadata and structure.
- Line 1516: Update the thumbnail entry in the image comparison configuration to
use input/bold_outfit_woman.jpeg as the first frame, while retaining
output/image_flux2_klein_image_edit_9b_distilled.png as the second frame. Do not
use input/handbag_white.png, since it is only a reference image.
In `@templates/index.fa.json`:
- Line 1516: Update the thumbnail entry for the compareSlider variant in
templates/index.fa.json to reference the distinct -1.webp and -2.webp assets in
the two array positions, replacing the duplicated image path while preserving
the existing thumbnail structure.
In `@templates/index.pt-BR.json`:
- Around line 9-34: Localize the newly added INT8 template titles and
descriptions in the Portuguese catalog, including image_z_image_int8,
image_boogu_image_0_1_edit_int8, and image_qwen_image_edit_2511_int8, plus the
identified new “Casos de Uso” entries. Replace their English user-facing strings
with Portuguese translations while preserving all metadata and non-text fields.
- Line 15743: Correct the title value in the relevant template entry from “Bira:
Remover Fundo da Imagem” to “Bria: Remover Fundo da Imagem”, keeping the
existing capitalization and surrounding translation unchanged.
- Line 14459: Remove the duplicated "Geração de Texto" value from the tags array
for this multimodal Gemma4 entry, replacing it with the intended distinct
image/vision category if that category is defined elsewhere in the template;
otherwise retain only one text-generation tag.
- Line 1516: Update the thumbnail array for the image edit entry in
templates/index.pt-BR.json by replacing its first duplicated
output/image_flux2_klein_image_edit_9b_distilled.png value with the
corresponding input asset path, while keeping the second output thumbnail
unchanged so the compare view displays before versus after.
In `@templates/index.ru.json`:
- Around line 11-12: Translate the title and description fields for the three
INT8 entries image_z_image_int8, image_boogu_image_0_1_edit_int8, and
image_qwen_image_edit_2511_int8 in the Russian catalog. Match the Russian
terminology and style used by their corresponding non-INT8 sibling entries,
while preserving the existing model and workflow details.
---
Outside diff comments:
In `@templates/index.fa.json`:
- Around line 1405-1411: Update the api_kling_o3_image entry’s thumbnail field
to reference the bundled preview asset api_kling_o3_image-1.webp instead of
output/api_kling_o3_image.json, while leaving the surrounding workflow metadata
unchanged.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 1a3e4d9f-3618-46d0-87c1-1e48d6fed057
⛔ Files ignored due to path filters (5)
input/giant_tulip_wanderer.pngis excluded by!**/*.pnginput/lily_girl_blue.pngis excluded by!**/*.pngoutput/image_boogu_image_0_1_edit_int8.pngis excluded by!**/*.pngoutput/image_qwen_image_edit_2511_int8.pngis excluded by!**/*.pngoutput/image_z_image_int8.pngis excluded by!**/*.png
📒 Files selected for processing (46)
archived/archived_i18n.jsonarchived/index.ar.jsonarchived/index.es.jsonarchived/index.fa.jsonarchived/index.fr.jsonarchived/index.ja.jsonarchived/index.jsonarchived/index.ko.jsonarchived/index.pt-BR.jsonarchived/index.ru.jsonarchived/index.tr.jsonarchived/index.zh-TW.jsonarchived/index.zh.jsonbundles.jsonpackages/core/src/comfyui_workflow_templates_core/manifest.jsonscripts/data/i18n.jsontemplates/api_bytedance_seedream_5_0_pro_image_edit.jsontemplates/api_bytedance_seedream_5_0_pro_t2i.jsontemplates/image_boogu_image_0_1_edit.jsontemplates/image_boogu_image_0_1_edit_int8-1.webptemplates/image_boogu_image_0_1_edit_int8-2.webptemplates/image_boogu_image_0_1_edit_int8.jsontemplates/image_qwen_image_edit_2509.jsontemplates/image_qwen_image_edit_2511.jsontemplates/image_qwen_image_edit_2511_int8-1.webptemplates/image_qwen_image_edit_2511_int8-2.webptemplates/image_qwen_image_edit_2511_int8.jsontemplates/image_z_image.jsontemplates/image_z_image_int8-1.webptemplates/image_z_image_int8.jsontemplates/index.ar.jsontemplates/index.es.jsontemplates/index.fa.jsontemplates/index.fr.jsontemplates/index.ja.jsontemplates/index.jsontemplates/index.ko.jsontemplates/index.mcp.jsontemplates/index.pt-BR.jsontemplates/index.ru.jsontemplates/index.tr.jsontemplates/index.zh-TW.jsontemplates/index.zh.jsontemplates/video_wan_dancer-1.webptemplates/video_wan_dancer.jsonworkflow_template_input_files.json
💤 Files with no reviewable changes (13)
- archived/index.zh-TW.json
- archived/index.zh.json
- archived/index.ja.json
- archived/index.fa.json
- archived/index.json
- archived/index.ar.json
- archived/index.tr.json
- archived/index.pt-BR.json
- archived/index.es.json
- archived/index.ru.json
- archived/index.ko.json
- archived/index.fr.json
- archived/archived_i18n.json
🛑 Comments failed to post (19)
bundles.json (1)
574-579: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash set -eu ids="$(jq -r '.["media-assets-01"][]' bundles.json)" for id in $ids; do count="$(rg -F "\"$id\"" packages/core/src/comfyui_workflow_templates_core/manifest.json | wc -l)" [ "$count" -eq 1 ] || echo "$id: expected exactly one manifest entry, found $count" doneRepository: Comfy-Org/workflow_templates
Length of output: 229
🏁 Script executed:
#!/bin/bash set -eu echo "== bundles.json context ==" nl -ba bundles.json | sed -n '566,584p' echo echo "== manifest references for media-assets-01 ids ==" python3 - <<'PY' import json, pathlib, re bundle = json.loads(pathlib.Path("bundles.json").read_text())["media-assets-01"] manifest = pathlib.Path("packages/core/src/comfyui_workflow_templates_core/manifest.json").read_text() for id in bundle: c = len(re.findall(re.escape(f'"{id}"'), manifest)) print(f"{id}: {c}") PY echo echo "== search for video_wan_dancer across repo ==" rg -n "video_wan_dancer" . echo echo "== inspect bundle-alignment test ==" fd -a "test_manifest_aligns_with_bundles_config" .Repository: Comfy-Org/workflow_templates
Length of output: 1688
🏁 Script executed:
#!/bin/bash set -eu echo "== locate bundle alignment test ==" rg -n "manifest_aligns_with_bundles_config|bundles_config|bundle.*align" . echo echo "== inspect package manifest shape around video_wan_dancer references ==" python3 - <<'PY' import json, pathlib p = pathlib.Path("packages/core/src/comfyui_workflow_templates_core/manifest.json") data = json.loads(p.read_text()) print(type(data).__name__) if isinstance(data, dict): print("top-level keys:", list(data)[:20]) # show any keys that mention "video" for k,v in data.items(): if "video_wan_dancer" in str(v): print("found under key:", k) PY echo echo "== inspect templates index entry ==" python3 - <<'PY' import json, pathlib p = pathlib.Path("templates/index.json") data = json.loads(p.read_text()) for item in data: if item.get("name") == "video_wan_dancer": print(item) break PY echo echo "== inspect bundles.json nearby lines with plain sed ==" sed -n '570,578p' bundles.jsonRepository: Comfy-Org/workflow_templates
Length of output: 1250
🏁 Script executed:
#!/bin/bash set -eu echo "== test_manifest_aligns_with_bundles_config ==" sed -n '1,160p' packages/core/tests/test_bundles_config.py echo echo "== manifest membership for video_wan_dancer ==" python3 - <<'PY' import json, pathlib manifest = json.loads(pathlib.Path("packages/core/src/comfyui_workflow_templates_core/manifest.json").read_text()) target = "video_wan_dancer" def walk(obj, path="$"): if isinstance(obj, dict): for k, v in obj.items(): if k == target or v == target: print(path, "->", k, "=", v if isinstance(v, (str, int, float, bool)) else type(v).__name__) walk(v, f"{path}.{k}") elif isinstance(obj, list): for i, v in enumerate(obj): if v == target: print(f"{path}[{i}] = {v}") walk(v, f"{path}[{i}]") walk(manifest) print("\nmanifest top-level keys:", list(manifest.keys())) PYRepository: Comfy-Org/workflow_templates
Length of output: 2392
🏁 Script executed:
#!/bin/bash set -eu echo "== loader.iter_templates source ==" sed -n '1,260p' packages/core/src/comfyui_workflow_templates_core/loader.py echo echo "== manifest.json template IDs ==" python3 - <<'PY' import json, pathlib manifest = json.loads(pathlib.Path("packages/core/src/comfyui_workflow_templates_core/manifest.json").read_text()) templates = manifest.get("templates", []) print("template_count:", len(templates)) for entry in templates: if isinstance(entry, dict) and entry.get("template_id") == "video_wan_dancer": print("FOUND:", entry) break else: print("FOUND: no") PYRepository: Comfy-Org/workflow_templates
Length of output: 8437
Add
video_wan_dancerto the manifest.It’s present in
bundles.jsonbut missing frompackages/core/src/comfyui_workflow_templates_core/manifest.json, sotest_manifest_aligns_with_bundles_configwill fail and the asset won’t resolve at runtime.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@bundles.json` around lines 574 - 579, Add the existing video_wan_dancer entry from bundles.json to packages/core/src/comfyui_workflow_templates_core/manifest.json, preserving the manifest’s established structure and ordering so test_manifest_aligns_with_bundles_config passes and the asset resolves at runtime.scripts/data/i18n.json (1)
793-796: 🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash # Verify video_wan_dancer: present in pending_templates AND has non-placeholder translations. fd -t f 'i18n.json' scripts/data --exec python3 - {} <<'PY' import json, sys p = sys.argv[1] data = json.load(open(p, encoding='utf-8')) pend = data.get("_status", {}).get("pending_templates", {}) tpl = data.get("templates", {}).get("video_wan_dancer", {}) print("in pending_templates:", "video_wan_dancer" in pend) if "video_wan_dancer" in pend: print("pending record:", pend["video_wan_dancer"]) title = tpl.get("title", {}) # flag any language whose value equals the English one (placeholder heuristic) en = title.get("en") dupes = [k for k,v in title.items() if k!="en" and v==en] print("title langs:", sorted(title.keys())) print("title langs mirroring 'en' (placeholder suspects):", dupes) PYRepository: Comfy-Org/workflow_templates
Length of output: 166
🏁 Script executed:
#!/bin/bash set -euo pipefail python3 - <<'PY' import json from pathlib import Path p = Path("scripts/data/i18n.json") data = json.loads(p.read_text(encoding="utf-8")) pend = data.get("_status", {}).get("pending_templates", {}) tpl = data.get("templates", {}).get("video_wan_dancer", {}) print("pending entry:", pend.get("video_wan_dancer")) print("pending keys:", list(pend.keys()) if "video_wan_dancer" in pend else "absent") for field in ("title", "description"): vals = tpl.get(field, {}) print(field, "langs:", sorted(vals.keys())) print(field, "en:", vals.get("en")) print(field, "unique values:", len(set(vals.values()))) PYRepository: Comfy-Org/workflow_templates
Length of output: 10234
Remove
video_wan_dancerfrom_status.pending_templatesIt already has localized
titleanddescriptionentries in every language, so the pending inventory is stale. Either drop it here or make the strings placeholders if it should still count as incomplete.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@scripts/data/i18n.json` around lines 793 - 796, Remove the video_wan_dancer entry from the _status.pending_templates inventory in scripts/data/i18n.json, since its title and description are already localized for all required languages. Do not alter the completed translation entries.templates/image_boogu_image_0_1_edit.json (1)
347-369: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
Fix mismatched markdown bold markers in the help note.
The
widgets_valuestext containsTry Boogu-Image-0.1-Edit-Turbo**— the**after "Turbo" is an unopened bold closer, so the markdown renderer will either show literal asterisks or mangle the formatting. It should be**Boogu-Image-0.1-Edit-Turbo**(wrapping the model name) or plainBoogu-Image-0.1-Edit-Turbowithout asterisks.✏️ Proposed fix
"## Faster Alternative\n\nNeed speed? Try Boogu-Image-0.1-Edit-Turbo** — 4-step distilled variant (~1K res).\n\n## Limitations\n\n + "## Faster Alternative\n\nNeed speed? Try **Boogu-Image-0.1-Edit-Turbo** — 4-step distilled variant (~1K res).\n\n## Limitations\n\n📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements."id": 55, "type": "MarkdownNote", "pos": [ -920, 4530 ], "size": [ 550, 1120 ], "flags": {}, "order": 3, "mode": 0, "inputs": [], "outputs": [], "title": "Note: Boogu Image Edit", "properties": {}, "widgets_values": [ "## Boogu-Image-0.1-Edit\n\n[Boogu/Boogu-Image-0.1-Edit](https://huggingface.co/Boogu/Boogu-Image-0.1-Edit) — a 10B Apache-2.0 unified image editing model (TI2I). Describe the change you want in natural language; the model keeps the subject coherent while editing backgrounds, objects, styles, text, and more.\n\n**Model page:** \n- [huggingface.co/Boogu/Boogu-Image-0.1-Edit](https://huggingface.co/Boogu/Boogu-Image-0.1-Edit)\n- [Comfy-Org/Boogu-Image](https://huggingface.co/Comfy-Org/Boogu-Image)\n\n## How to Use\n\n1. Load a reference image in **Load Image** (one image only for now).\n2. Write an editing instruction in the subgraph prompt, e.g. *\"Keep the character unchanged, replace the desert background and scene. The model is on the dune.\"*\n3. Run the workflow. Use **Image Compare** to check before/after.\n\n## Resolution\n\nSupported: **1K, 1.5K, 2K**. Results are most stable at **1K**; use the Resize node megapixels setting for higher res (up to ~4 MP depending on VRAM). \n## What It Can Do\n\n- Object insert / replace / remove\n- Attribute & material changes\n- Background & scene replacement\n- Style transfer (e.g. *\"Change the style to a colored pencil drawing.\"*)\n- Bilingual (CN/EN) in-image text editing\n\n## Faster Alternative\n\nNeed speed? Try **Boogu-Image-0.1-Edit-Turbo** — 4-step distilled variant (~1K res).\n\n## Limitations\n\n- Single reference image only (multi-image coming later)\n- Strict identity/layout preservation can be unstable on hard edits\n- Long or dense text edits may drift; small faces/limbs can artifact (FLUX VAE)\n" ], "color": "`#222`", "bgcolor": "`#000`" }🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/image_boogu_image_0_1_edit.json` around lines 347 - 369, Update the MarkdownNote node’s widgets_values text in “Faster Alternative” so the Boogu-Image-0.1-Edit-Turbo model name has balanced Markdown bold markers, preferably wrapping the complete name with ** on both sides. Leave the surrounding help note content unchanged.templates/image_qwen_image_edit_2509.json (1)
129-133: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash # Inspect the subgraph input defs and their linkIds for lora-related ports. fd 'image_qwen_image_edit_2509.json' templates --exec \ python3 -c ' import json,sys d=json.load(open(sys.argv[1])) sg=d["definitions"]["subgraphs"][0] for i,inp in enumerate(sg["inputs"]): if "lora" in inp["name"]: print(i, inp["name"], "linkIds=", inp.get("linkIds")) ' {}Repository: Comfy-Org/workflow_templates
Length of output: 220
🏁 Script executed:
#!/bin/bash set -euo pipefail # Map the relevant JSON structure in the target template and its sibling. for f in templates/image_qwen_image_edit_2509.json templates/image_qwen_image_edit_2511.json; do echo "### $f" python3 - <<'PY' "$f" import json, sys path = sys.argv[1] d = json.load(open(path)) sg = d["definitions"]["subgraphs"][0] print("subgraph inputs:") for i, inp in enumerate(sg["inputs"]): if "lora" in inp["name"] or i in (4, 11): print(i, inp["name"], "linkIds=", inp.get("linkIds")) print("top-level node inputs:") for node in d["nodes"]: if node.get("type") == "subgraph" or node.get("name") == "subgraph": pass for node in d.get("nodes", []): if node.get("type") == "Subgraph" or node.get("type") == "subgraph" or node.get("name") == "subgraph": print(node.get("id"), node.get("type"), node.get("inputs", [])) PY echo done # Find any references to slot 4 / lora_name in the target template. python3 - <<'PY' import json d = json.load(open('templates/image_qwen_image_edit_2509.json')) hits = [] def walk(x, path=""): if isinstance(x, dict): for k,v in x.items(): if k == "name" and v == "lora_name": hits.append(path) walk(v, f"{path}/{k}") elif isinstance(x, list): for i,v in enumerate(x): walk(v, f"{path}[{i}]") walk(d) print("lora_name occurrences:", len(hits)) for h in hits[:20]: print(h) PYRepository: Comfy-Org/workflow_templates
Length of output: 625
Remove the orphaned
lora_nameinput —templates/image_qwen_image_edit_2509.jsonexposes slot 4 on the subgraph node, but its subgraph input has nolinkIds, so it’s a dead dial; the real LoRA path islora_name_1/link 742. Drop it (and reindex ports) unless it’s meant as a placeholder.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/image_qwen_image_edit_2509.json` around lines 129 - 133, Remove the orphaned lora_name input from the relevant subgraph node in image_qwen_image_edit_2509.json, then reindex subsequent input ports so the node no longer exposes the unused slot. Preserve the functional lora_name_1 input and its link 742.templates/image_qwen_image_edit_2511_int8.json (2)
438-448: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Label/behavior mismatch on the exposed toggle. This subgraph input is labeled
enable_turbo_mode, yet its link (385) drives node168titled "Enable 4steps LoRA?" and the CFG/steps/model switches. The name and the deed don't quite agree — rename the label to match (e.g.enable_4steps_lora) so users aren't left guessing which knob they're turning.🏷️ Proposed label fix
"type": "BOOLEAN", "linkIds": [ 385 ], - "label": "enable_turbo_mode", + "label": "enable_4steps_lora",📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements."id": "0a8c3d7a-76e3-487c-8957-b4fb68b443e5", "name": "value", "type": "BOOLEAN", "linkIds": [ 385 ], "label": "enable_4steps_lora", "pos": [ -1484.255859375, 3134 ]🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/image_qwen_image_edit_2511_int8.json` around lines 438 - 448, Rename the exposed BOOLEAN input label from enable_turbo_mode to enable_4steps_lora, keeping its link ID 385 and existing behavior unchanged.
1836-1836: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Typo in group title. "Ligtning" should be "Lightning" — a stray letter took flight. This label is user-visible in the graph.
✍️ Proposed fix
- "title": "4 Steps Ligtning LoRA", + "title": "4 Steps Lightning LoRA",📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements."title": "4 Steps Lightning LoRA",🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/image_qwen_image_edit_2511_int8.json` at line 1836, Correct the user-visible group title in the template by changing “4 Steps Ligtning LoRA” to “4 Steps Lightning LoRA”.templates/image_qwen_image_edit_2511.json (1)
1276-1276: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
Default UNET (
fp8mixed) doesn't match the model note.The loader now defaults to
qwen_image_edit_2511_fp8mixed.safetensors, but the "Note: Model Links" block (Line 26) only lists/storesqwen_image_edit_2511_bf16.safetensors. Users who follow the note will grab bf16 and wonder why the graph reaches for a file they never fetched — a fp8-faux-pas worth smoothing over. Please add thefp8mixedentry to the note (or realign the default) so the docs and the download match.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/image_qwen_image_edit_2511.json` at line 1276, Update the “Note: Model Links” block to include qwen_image_edit_2511_fp8mixed.safetensors alongside the existing model entry, keeping the default UNET filename in the loader aligned with the documented download.templates/image_z_image.json (1)
341-341: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Subgraph name says "Int8" but this is the bf16 base graph.
The subgraph is named
"Text to Image(Z-Image-Base Int8)", yet the UNETLoader here loadsz_image_bf16.safetensors(not an INT8 build). That "Int8" tag is a bit of a bit-flip — drop it to avoid confusing this base template with the separateimage_z_image_int8one.✏️ Proposed fix
- "name": "Text to Image(Z-Image-Base Int8)", + "name": "Text to Image(Z-Image-Base)",📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements."name": "Text to Image(Z-Image-Base)",🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/image_z_image.json` at line 341, Update the subgraph name in the relevant template metadata from “Text to Image(Z-Image-Base Int8)” to identify the bf16 base graph, removing the “Int8” designation while leaving the UNETLoader configuration and separate INT8 template unchanged.templates/index.ar.json (1)
14459-14459: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash set -euo pipefail # Inspect the localized file around the reported line and locate the matching source entry. git ls-files 'templates/index.ar.json' 'templates/index.json' | sed -n '1,20p' printf '\n--- localized context ---\n' sed -n '14445,14470p' templates/index.ar.json printf '\n--- source search ---\n' rg -n '"llm_gemma4_text_gen"|توليد النص' templates/index.json templates/index.ar.jsonRepository: Comfy-Org/workflow_templates
Length of output: 2089
🏁 Script executed:
#!/bin/bash set -euo pipefail printf -- '--- source context ---\n' sed -n '14445,14470p' templates/index.jsonRepository: Comfy-Org/workflow_templates
Length of output: 1074
Remove the duplicate
"توليد النص"tagOne tag is enough here. The same duplicate exists in
templates/index.json, so fix the source entry and regenerate this mirror.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.ar.json` at line 14459, Remove the duplicate "توليد النص" value from the tags entry in the source template, leaving a single tag, then regenerate the mirrored templates/index.ar.json and templates/index.json entries so both outputs stay consistent.templates/index.es.json (4)
11-12: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Untranslated strings in the Spanish (
.es.json) catalog.These freshly-added Int8 entries wear their English coat in a Spanish closet. "Z-Image Int8: Text to Image" and the Boogu description remain in English while every neighbouring entry is localized (e.g.,
image_z_image→ "Z-Image: Texto a Imagen"). Please translatetitle/descriptionto keep the UI consistent for Spanish users.🌐 Suggested localization
- "title": "Z-Image Int8: Text to Image", - "description": "Generate high-quality, diverse images from text prompts using the Z-Image Int8 model, which leverages an undistilled transformer for precise prompt adherence and broad stylistic coverage. The workflow takes no file inputs and produces 1 image output through a standard text-to-image pipeline. Ideal for professional creative projects, LoRA training and model development, and scenarios requiring robust negative control and high compositional diversity.", + "title": "Z-Image Int8: Texto a Imagen", + "description": "Genere imágenes diversas y de alta calidad a partir de indicaciones de texto usando el modelo Z-Image Int8, ...",As per path instructions, when catalog content changes, keep affected references consistent across workflows and localized files.
Also applies to: 37-38
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.es.json` around lines 11 - 12, Translate the English title and description for the newly added Z-Image Int8 and Boogu entries in the Spanish catalog into natural Spanish, matching neighboring localized terminology. Update all affected catalog references consistently across workflows and localized files while preserving the existing metadata and structure.Source: Path instructions
241-242: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Another English stowaway in the Spanish catalog.
"Qwen Image Edit 2511 Int8: Image Edit" and its description are in English. Translate to match the localized siblings (e.g., "Qwen Image Edit 2511 - Reemplazo de Materiales").
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.es.json` around lines 241 - 242, Translate the title and description for the Qwen Image Edit 2511 Int8 catalog entry into Spanish, matching the terminology and style of nearby localized entries such as “Qwen Image Edit 2511 - Reemplazo de Materiales.” Preserve the model name, quantization details, input/output behavior, use cases, and JSON structure.
1516-1516: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash set -euo pipefail echo "== File excerpt ==" sed -n '1490,1535p' templates/index.es.json | cat -n echo echo "== Search for this template id / nearby entries ==" rg -n '"thumbnailVariant"|image_flux2_klein_image_edit_9b_distilled|compareSlider' templates/index.es.json echo echo "== Search for matching input/output assets in repo ==" fd -a 'image_flux2_klein_image_edit_9b_distilled' .Repository: Comfy-Org/workflow_templates
Length of output: 11407
🏁 Script executed:
#!/bin/bash set -euo pipefail echo "== compareSlider thumbnail examples ==" python3 - <<'PY' import re, json from pathlib import Path text = Path("templates/index.es.json").read_text() # crude scan for entries with compareSlider and thumbnail arrays for m in re.finditer(r'"thumbnailVariant"\s*:\s*"compareSlider"', text): start = text.rfind('{', 0, m.start()) end = text.find('}', m.end()) PYRepository: Comfy-Org/workflow_templates
Length of output: 205
🏁 Script executed:
#!/bin/bash set -euo pipefail python3 - <<'PY' import json from pathlib import Path data = json.loads(Path("templates/index.es.json").read_text()) print(type(data).__name__) if isinstance(data, dict): print(data.keys()) # show first few top-level items if list under a key for k,v in data.items(): if isinstance(v, list): print(k, len(v)) break elif isinstance(data, list): print("len", len(data)) PYRepository: Comfy-Org/workflow_templates
Length of output: 177
🏁 Script executed:
#!/bin/bash set -euo pipefail python3 - <<'PY' import json from pathlib import Path data = json.loads(Path("templates/index.es.json").read_text()) print(type(data).__name__) if isinstance(data, dict): print(list(data.keys())) for k,v in data.items(): if isinstance(v, list): print(k, len(v)) break elif isinstance(data, list): print("len", len(data)) PYRepository: Comfy-Org/workflow_templates
Length of output: 177
🏁 Script executed:
#!/bin/bash set -euo pipefail echo "== Template JSON for the entry ==" sed -n '1,220p' templates/image_flux2_klein_image_edit_9b_distilled.json | cat -n echo echo "== Search for any non-output matching image assets ==" fd -a 'image_flux2_klein_image_edit_9b_distilled*' templates output . echo echo "== Search for input/image references in nearby template metadata ==" rg -n '"thumbnail"|compareSlider|input/|output/' templates/image_flux2_klein_image_edit_9b_distilled.json templates/index.es.jsonRepository: Comfy-Org/workflow_templates
Length of output: 48066
🏁 Script executed:
#!/bin/bash set -euo pipefail echo "== Search for referenced input assets ==" fd -a 'bold_outfit_woman.jpeg' . || true fd -a 'handbag_white.png' . || true echo echo "== Search for the template's exact image names in thumbnails/inputs/outputs ==" fd -a 'image_flux2_klein_image_edit_9b_distilled' templates input output thumbnail . || true echo echo "== Nearby template metadata around the entry ==" sed -n '1468,1522p' templates/index.es.json | cat -nRepository: Comfy-Org/workflow_templates
Length of output: 3435
🏁 Script executed:
#!/bin/bash set -euo pipefail python3 - <<'PY' import json from pathlib import Path data = json.loads(Path("templates/image_flux2_klein_image_edit_9b_distilled.json").read_text()) nodes = {n["id"]: n for n in data["nodes"]} for nid in (76, 121, 75, 9): n = nodes[nid] print(f"NODE {nid} {n['type']}") print(" widgets_values:", n.get("widgets_values")) for inp in n.get("inputs", []): print(" input", inp) for out in n.get("outputs", []): print(" output", out) print() print("Connections into node 75:") for inp in nodes[75]["inputs"]: if inp.get("link") is not None: link = inp["link"] # find source for n in data["nodes"]: for out in n.get("outputs", []): if link in (out.get("links") or []): print(f" link {link} from node {n['id']} {n['type']} output {out['name']}") PYRepository: Comfy-Org/workflow_templates
Length of output: 1016
🏁 Script executed:
#!/bin/bash set -euo pipefail python3 - <<'PY' import json from pathlib import Path data = json.loads(Path("templates/image_flux2_klein_image_edit_9b_distilled.json").read_text()) nodes = data["nodes"] # Find which node consumes link 189 from node 121. consumers = [] for n in nodes: for inp in n.get("inputs", []): if inp.get("link") == 189: consumers.append((n["id"], n["type"], inp.get("name"), inp.get("label"))) print("consumers of link 189:", consumers) # Show a compact neighborhood of node 121 and linked consumers. for nid in [121] + [c[0] for c in consumers]: n = next(x for x in nodes if x["id"] == nid) print(f"\nNODE {nid} {n['type']}") print(" widgets_values:", n.get("widgets_values")) print(" inputs:", n.get("inputs")) print(" outputs:", n.get("outputs")) PYRepository: Comfy-Org/workflow_templates
Length of output: 859
🏁 Script executed:
#!/bin/bash set -euo pipefail python3 - <<'PY' import json from pathlib import Path data = json.loads(Path("templates/image_flux2_klein_image_edit_9b_distilled.json").read_text()) nodes = data["nodes"] consumers = [] for n in nodes: for inp in n.get("inputs", []): if inp.get("link") == 189: consumers.append((n["id"], n["type"], inp.get("name"), inp.get("label"))) print("consumers of link 189:", consumers) for nid in [121] + [c[0] for c in consumers]: n = next(x for x in nodes if x["id"] == nid) print(f"\nNODE {nid} {n['type']}") print(" widgets_values:", n.get("widgets_values")) print(" inputs:", n.get("inputs")) print(" outputs:", n.get("outputs")) PYRepository: Comfy-Org/workflow_templates
Length of output: 859
Use the edit source image for the compare slider
thumbnailis duplicated right now, so the slider is a no-op. Swap the first frame toinput/bold_outfit_woman.jpeg;input/handbag_white.pngis only a reference image.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.es.json` at line 1516, Update the thumbnail entry in the image comparison configuration to use input/bold_outfit_woman.jpeg as the first frame, while retaining output/image_flux2_klein_image_edit_9b_distilled.png as the second frame. Do not use input/handbag_white.png, since it is only a reference image.
8931-8931: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Two more entries speaking English at a Spanish party.
In the new "Casos de Uso" section,
template_seedance2_0_viral_videos_character_swapandtemplate_qwen_image_edit_2511_systms_actioncarry Englishdescription(and the former an Englishtitle). Localize them to keep the essential section coherent for Spanish users.Also applies to: 9344-9344
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.es.json` at line 8931, Localize the English title and description for template_seedance2_0_viral_videos_character_swap and the description for template_qwen_image_edit_2511_systms_action within the new “Casos de Uso” section of templates/index.es.json. Preserve each entry’s meaning while translating the user-facing text into Spanish.templates/index.fa.json (1)
1516-1516: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash # Look for candidate thumbnail assets for this template fd -a 'image_flux2_klein_image_edit_9b_distilled' templates 2>/dev/null || true rg -n 'image_flux2_klein_image_edit_9b_distilled' --glob '!templates/index.*.json' -lRepository: Comfy-Org/workflow_templates
Length of output: 398
🏁 Script executed:
#!/bin/bash set -euo pipefail # Inspect the surrounding entry in the localized catalog sed -n '1470,1525p' templates/index.fa.json | cat -n # Show the actual asset files that match this template fd -a 'image_flux2_klein_image_edit_9b_distilled' templates 2>/dev/null || true # Check whether the two referenced thumbnail variants exist and what the catalog points to nearby python3 - <<'PY' import json from pathlib import Path p = Path("templates/index.fa.json") data = json.loads(p.read_text(encoding="utf-8")) for i, item in enumerate(data): if item.get("name") == "image_flux2_klein_image_edit_9b_distilled": print("index:", i) print("thumbnailVariant:", item.get("thumbnailVariant")) print("thumbnail:", item.get("thumbnail")) break PYRepository: Comfy-Org/workflow_templates
Length of output: 2956
🏁 Script executed:
#!/bin/bash set -euo pipefail # Find other compareSlider entries and inspect their thumbnail paths python3 - <<'PY' import json from pathlib import Path data = json.loads(Path("templates/index.fa.json").read_text(encoding="utf-8")) count = 0 for item in data: if item.get("thumbnailVariant") == "compareSlider": print("NAME:", item.get("name")) print("THUMBNAIL:", item.get("thumbnail")) count += 1 if count >= 8: break PYRepository: Comfy-Org/workflow_templates
Length of output: 166
Use distinct thumbnails for the compare slider
templates/index.fa.json:1516—thumbnailVariant: "compareSlider"points both slots at the same image, so the before/after view collapses into a single frame. Point them at the-1.webpand-2.webpassets instead; two copies of the same file make the slider a snoozer.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.fa.json` at line 1516, Update the thumbnail entry for the compareSlider variant in templates/index.fa.json to reference the distinct -1.webp and -2.webp assets in the two array positions, replacing the duplicated image path while preserving the existing thumbnail structure.templates/index.pt-BR.json (4)
9-34: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
New INT8 entries carry English strings in a Portuguese catalog.
This is the pt-BR localized catalog, yet the newly added INT8 templates ship English titles/descriptions while their neighbors speak Portuguese. For example, "Z-Image Int8: Text to Image" plus its English description here, the same at Lines 37-38 (
image_boogu_image_0_1_edit_int8), and Lines 241-242 (image_qwen_image_edit_2511_int8, title"...Int8: Image Edit"). A few new "Casos de Uso" entries share the slip (e.g., Lines 8931, 9344). Please localize these strings so the catalog doesn't switch tongues mid-scroll — a little translation keeps the flow in fashion.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.pt-BR.json` around lines 9 - 34, Localize the newly added INT8 template titles and descriptions in the Portuguese catalog, including image_z_image_int8, image_boogu_image_0_1_edit_int8, and image_qwen_image_edit_2511_int8, plus the identified new “Casos de Uso” entries. Replace their English user-facing strings with Portuguese translations while preserving all metadata and non-text fields.
1516-1516: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win
Duplicate thumbnail path — the slider is comparing a frame with itself. Replace the first thumbnail entry with the matching input asset so the compare view shows before vs. after.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.pt-BR.json` at line 1516, Update the thumbnail array for the image edit entry in templates/index.pt-BR.json by replacing its first duplicated output/image_flux2_klein_image_edit_9b_distilled.png value with the corresponding input asset path, while keeping the second output thumbnail unchanged so the compare view displays before versus after.
14459-14459: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Duplicated tag value — "Geração de Texto" twice is one too nice.
"tags": ["Geração de Texto", "Geração de Texto"] repeats the same tag; the second was likely meant to be another category (e.g., an image/vision tag, given this multimodal Gemma4 entry).
🏷️ Proposed fix
- "tags": ["Geração de Texto", "Geração de Texto"], + "tags": ["Geração de Texto"],📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements."tags": ["Geração de Texto"],🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.pt-BR.json` at line 14459, Remove the duplicated "Geração de Texto" value from the tags array for this multimodal Gemma4 entry, replacing it with the intended distinct image/vision category if that category is defined elsewhere in the template; otherwise retain only one text-generation tag.
15743-15743: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Typo in title: "Bira" should be "Bria".
The
name,models, andproviderall sayBRIA, but the title reads "Bira: Remover Fundo da Imagem". One transposed vowel, easy to repair.✏️ Proposed fix
- "title": "Bira: Remover Fundo da Imagem", + "title": "Bria: Remover Fundo da Imagem",📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements."title": "Bria: Remover Fundo da Imagem",🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.pt-BR.json` at line 15743, Correct the title value in the relevant template entry from “Bira: Remover Fundo da Imagem” to “Bria: Remover Fundo da Imagem”, keeping the existing capitalization and surrounding translation unchanged.templates/index.ru.json (1)
11-12: 📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win
Untranslated English strings in the Russian catalog.
This is
index.ru.json, yet the three new INT8 entries ship English titles/descriptions — a small localization slip that leaves RU users lost in translation:
- Lines 11–12 (
image_z_image_int8)- Lines 37–38 (
image_boogu_image_0_1_edit_int8)- Lines 241–242 (
image_qwen_image_edit_2511_int8)Please supply the Russian equivalents (matching how the sibling
image_z_image/image_qwen_image_edit_2511entries are localized).🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@templates/index.ru.json` around lines 11 - 12, Translate the title and description fields for the three INT8 entries image_z_image_int8, image_boogu_image_0_1_edit_int8, and image_qwen_image_edit_2511_int8 in the Russian catalog. Match the Russian terminology and style used by their corresponding non-INT8 sibling entries, while preserving the existing model and workflow details.Source: Coding guidelines
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.