User repository on GitHub
Jev text interpretation and judgments for ComfyUI.
- Forks
- 1
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Integrations & Adapters
ComfyUI-Jev
Custom ComfyUI nodes that use Jev for text interpretation and judgments — select candidates, evaluate conditions, score text and extract numbers — plus a node for picking local SKILL.md files.
- Stars
- 7
- Language
- Python
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- hndrr
- Last verified
- Oct 2026
- Languages
- Python
- Platforms
- OpenRouter
- Capabilities
- Open sourceRequires API key
Highlights
- Three nodes: Jev Interpret (select, judge, score, extract numbers), Jev Skill Choice (pick local SKILL.md files) and OpenRouter Text.
- Requires ComfyUI v0.36.0 or later and Python 3.10 or later; installs via ComfyUI Manager or a git clone into custom_nodes.
- Judgments use the TypeSafe API by default; the provider can switch to OpenRouter, with keys from the environment or a per-node field.
- Keys entered in node fields are saved in workflows and execution history, so the README says to remove them before sharing.
- Image inputs are optional; image-capable Decisions models such as Luna, Clef and Clef Flash pair with a Load Image node.
Quickstart
bash
cd /path/to/ComfyUI/custom_nodes
git clone https://github.com/hndrr/ComfyUI-Jev.gitWatch out
Copy reviewed against the repo README on 2026-10-15; the README is marked WORK IN PROGRESS and was truncated after the decisions-with-images section, so the node reference and example workflows are not covered. A package entry exists on registry.comfy.org.
Questions about this entry
Short answers drawn from the linked project or page.
A set of custom ComfyUI nodes for using Jev's text interpretation and judgments inside workflows: selecting candidates, evaluating conditions, scoring text and extracting numbers with natural-language instructions, then passing the results to other nodes.
Three: Jev Interpret, which selects, judges, scores or extracts numbers; Jev Skill Choice, which selects local SKILL.md files for a request; and OpenRouter Text, which generates text or selection candidates with OpenRouter.
It requires ComfyUI v0.36.0 or later and Python 3.10 or later. Either search for ComfyUI-Jev in ComfyUI Manager and install it, or git clone the repository into your ComfyUI custom_nodes directory, then restart ComfyUI.
Jev Interpret and Jev Skill Choice use TypeSafe by default, so set TYPESAFE_API_KEY in the environment that starts ComfyUI, or enter a key in the node. OpenRouter Text, and optionally the judgment nodes, need OPENROUTER_API_KEY.
If you type a key into a node's api_key field, it is saved in workflows and execution history; the README advises removing such keys before sharing a workflow and using environment variables instead. An explicit key takes precedence over the environment.
The images input is optional and both judgment nodes work with it disconnected. To add images, select an image-capable Decisions model such as Luna, Clef or Clef Flash and connect a Load Image output to the images input.
The README is marked WORK IN PROGRESS and a package is published on registry.comfy.org. The nodes load only Decisions models from the public catalog and switch between providers through the provider setting.
More Jev resources like this
A dbt package that evaluates SQL values with Jev through TypeSafe's hosted API or OpenRouter: macros expose Choice as a nullable label, Noul as a match probability and Score as a numeric rating, on DuckDB and ClickHouse.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
smithclay/dbt_jevPython
smithclay/dbt_jevPythonA Python library that runs LLM and Jev calls as Polars expressions: parallel inference across providers, Pydantic-validated outputs as struct columns, and a native TypeSafe System One layer for per-row decision questions.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
pnthn-ai/polar_llamaPython
pnthn-ai/polar_llamaPythonA Python tool that asks TypeSafe Jev to fix Japanese readings in VOICEVOX: Sudachi splits sentences, Jev answers reading and phrase-boundary questions, and output comes as corrected WAV files or a VOICEVOX-compatible API server.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
nemalabs/voicevox-jev-proxyPython
nemalabs/voicevox-jev-proxyPythonAn independent agent skill that walks a coding agent through designing a bounded Jev decision layer: contract, criteria, error behavior, fallback, and evaluation plan first, then scaffolding through TypeSafe or OpenRouter.
Jev Agents & MCP#community#catalog-import#logicrw
Community
grandamenium/jev-anythingPython
grandamenium/jev-anythingPythonA GitHub App that routes pull-request review work through Jev: it ingests the webhook, normalizes the PR context, gets a LOW, NORMAL or SPECIALIST decision, and a deterministic policy layer maps it to auto-approval, a review request or a check-run update.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
hexyedev/jevprPython
hexyedev/jevprPythonA Playwright reporter that uses Jev to score which existing E2E tests a code change could affect and runs only those above a confidence threshold before browser execution, failing open to the full suite.
Jev Integrations & Adapters#community#catalog-import#logicrw
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.