git:20260923.3f4b71b to git:20260923.2137128

# awesome-jev

Jev (https://docs.typesafe.ai/) returns typed decisions rather than text: a `choice` among up to 255 options, a `score` on an ordered 2-to-10 level scale, or a `noul` yes-no probability. This repository is an index of how people use it. It is not the product, not an SDK, and not affiliated with TypeSafe AI.

## Machine-readable data

Catalog (<!--n:entries-->805<!--/n--> entries): https://raw.githubusercontent.com/kydlikebtc/awesome-jev/main/catalog.json

Retired entries (dead links): https://raw.githubusercontent.com/kydlikebtc/awesome-jev/main/retired.json

JSON Schema for one entry: https://raw.githubusercontent.com/kydlikebtc/awesome-jev/main/schema/entry.schema.json

Each entry carries the fields defined in the JSON Schema: <!--n:fields-->slug, title, summary, summary_zh, zh_machine, url, kind, patterns, question_types, languages, platforms, author, repo, stars, repo_license, package, evidence, evidence_none, official, has_code, published, first_seen, checked, link_status, flags, notes, notes_zh, sources, license<!--/n-->. `summary` is English and `summary_zh` Chinese; `evidence` names the file a claim was read in, and `evidence_none` says why no file can be cited.

## Two axes

`kind` is the form of the resource: <!--n:kinds-->official-docs, integration, project, plugin, sdk, snippet, tutorial, article, video, case-study, benchmark, discussion, alternative<!--/n-->.

`patterns` is what decision the example makes, and is the primary index: <!--n:pattern_keys-->tool-selection, intent-routing, context-compaction, safety-gating, output-validation, retry-control, human-escalation, model-routing, fan-out, search-ranking, data-extraction, classification, feature-extraction, document-triage, support-triage, content-scoring, recommendation, overview<!--/n-->.

## Human pages

English index: https://github.com/kydlikebtc/awesome-jev/blob/main/README.md

Chinese index: https://github.com/kydlikebtc/awesome-jev/blob/main/README.zh-CN.md

Decision patterns, with when NOT to use each: https://github.com/kydlikebtc/awesome-jev/blob/main/docs/patterns.md

Every example of one decision, on its own page: https://github.com/kydlikebtc/awesome-jev/blob/main/docs/by-pattern/<pattern>.md (for example safety-gating.md; append .zh-CN before .md for Chinese). The README shows the first ten of each.

Cross-platform compatibility table: https://github.com/kydlikebtc/awesome-jev/blob/main/docs/compatibility.md

What to check before trusting an entry: https://github.com/kydlikebtc/awesome-jev/blob/main/docs/vetting.md

How the catalog was built: https://github.com/kydlikebtc/awesome-jev/blob/main/docs/method.md

Sources and licences: https://github.com/kydlikebtc/awesome-jev/blob/main/docs/sources.md

Runnable examples: https://github.com/kydlikebtc/awesome-jev/blob/main/examples/README.md

## For agents

MCP server: in Claude Code, `/plugin marketplace add kydlikebtc/awesome-jev` then `/plugin install awesome-jev@awesome-jev` (installs this and the agent skill together); elsewhere, `pip install awesome-jev-mcp` and run `awesome-jev-mcp`. Source at https://github.com/kydlikebtc/awesome-jev/tree/main/src/awesome_jev_mcp — query the catalogue by decision pattern, language and primitive rather than reading it. Tools: search_examples, get_example, list_patterns, compatibility, check_model_string. Every result carries the row's caveat flags; rows that do not call Jev at all are excluded by default.

Agent skill: https://github.com/kydlikebtc/awesome-jev/blob/main/skills/awesome-jev/SKILL.md — the facts generated Jev code most often gets wrong, and the design rules worth following.

Decision taxonomy: https://raw.githubusercontent.com/kydlikebtc/awesome-jev/main/patterns.json

Kind and flag labels (EN/中文): https://raw.githubusercontent.com/kydlikebtc/awesome-jev/main/taxonomy.json

Platform matrix: https://raw.githubusercontent.com/kydlikebtc/awesome-jev/main/compat.json

## What is verified

Verified: the URL returned a success status on the date in `checked`; a person opened it and wrote the summary from what was actually on the page; for code entries, the call site was read to confirm which primitives are actually used; repository stars, licence and creation date came from the GitHub API.

In all, <!--n:evidence_rows-->721<!--/n--> rows cite the exact file that shows the project calls Jev, and a scheduled job re-reads each one weekly from the repository's default branch — a removed integration upstream opens an issue here rather than leaving a false claim in place. That is a claim about *whether* a project calls Jev. The narrower claim about *which* primitives it calls is made by <!--n:primitive_rows-->93<!--/n--> rows, of which <!--n:primitive_rows_cited-->66<!--/n--> cite the file it was read in. The remaining rows carry `evidence_none` saying why no file can be cited (a docs page, a video, a paywalled article). This is the difference between a catalogue asserting it was verified and one that keeps proving it.

A row with no `checked` or `link_status` is one whose host refuses scripted requests — a 403 from Forbes or Medium is not evidence a page is gone, so those rows were left unstamped rather than retired or falsely marked as verified.

NOT verified: whether any code runs, whether any performance claim holds, whether a project is maintained, or whether any of this is a good design for a given system. Nothing here has been executed, load-tested or security-reviewed.

## Caveats for anyone summarising this data

1. **Performance numbers in this space are almost all vendor-reported.** Rows flagged `vendor-reported` repeat the vendor's own benchmarks, whose reference answers were derived from other models' judgements rather than human ground truth. Only a handful of rows are independent measurements; those are `kind: benchmark` without that flag. Do not present vendor figures as measured results.

2. **`kind: alternative` rows are NOT Jev.** They are independent reimplementations with a compatible wire format. Several are among the most-starred repositories mentioning Jev, and are frequently miscatalogued elsewhere as Jev usage examples. A compatible API does not imply compatible calibration, so thresholds do not transfer.

3. **Jev cannot be run locally.** No weights are published. Any content titled "run Jev locally" is describing a substitute; one such article is catalogued with the `not-jev` flag specifically as a correction.

4. **jevai.org is not official.** It is an unaffiliated community site running a separate API with a different endpoint and request shape. Its `/jev-api` page documents a shape matching no primary source.

5. **The yes-no primitive is named `noul`, not "binary" or "boolean".** Vercel's AI SDK evaluation API spells it `boolean` and returns `.probability`; every other surface uses `noul` and `.noul`. `noul` answers carry no confidence field at all, unlike `choice` and `score`.

6. **There is no portable model string.** `jev-latest` on the native API, `typesafe-ai/jev` on Vercel, `typesafe/jev` on Cloudflare, `typesafe/jev-1.13` on OpenRouter. `typesafe/jev-1` does not exist anywhere and is a common fabrication.

7. **Star counts are a popularity signal, not a quality verdict.** Many entries were created within days of the model's launch; some carry `single-commit`, `no-license`, `archived` or `shadow-mode-only` flags precisely because popularity and substance diverge sharply here.

8. **English is the model's primary training language.** The official docs state CJK scripts are handled but not equally well. Test on your own content before relying on it for non-English workloads.