# AI use cases for Awesome Jev (kraayenjon/awesome-jev) Repo: [kraayenjon/awesome-jev](https://github.com/kraayenjon/awesome-jev) | Use cases: 4 | Page: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/ An unofficial curated list of Jev use cases, SDKs, tools, and learning resources for TypeSafe AI's System One model, which returns typed decisions instead of generated text. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Route support tickets by intent and urgency in Python: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/route-support-tickets-python/ - Try Jev routing rules in the browser Playground: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/test-jev-rules-in-playground/ - Add Jev decisions to a JavaScript or TypeScript app: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/add-jev-decisions-to-javascript-app/ - Give a coding agent Jev skills for decision steps: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/give-coding-agent-jev-skill/ ## How to route support tickets by intent and urgency in Python with Awesome Jev Repo: [kraayenjon/awesome-jev](https://github.com/kraayenjon/awesome-jev) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Sort incoming tickets into intent categories and flag urgent ones with the Python SDK. **Who it is for:** A developer on a support or operations team who wants tickets sorted before a person reads them, triggered by a growing backlog of plain-text messages. **The problem:** Keyword rules miss meaning, and sending every ticket to an LLM means parsing free-form text back into fields. **The outcome:** A script that returns one intent label and one urgency score per ticket, ready to branch on in code. ### Steps 1. Run `pip install typesafe-sdk`. 2. Create an API key at https://console.typesafe.ai/settings/keys and set it as the TYPESAFE_API_KEY environment variable. 3. Copy the Python quick start from the README, replace the sample state with your ticket text, and keep the Choice, Noul, and Score questions. 4. Run the script on a few sample tickets and check the intent, is_urgent, and frustration values. 5. Add your own thresholds in code to decide which tickets go to a priority queue. ### Prompt to paste into your agent ```text Using the repo kraayenjon/awesome-jev and the typesafe-sdk Python package, write a script that reads each ticket in and sends it to Jev with a Choice question for intent, a Noul question for urgency, and a Score question for frustration. Output one line per ticket with the intent, the urgency value, and the frustration score. Verify that every urgency value is between 0.0 and 1.0 and that every intent is one of the option keys you defined. ``` ### Caveat Jev accepts text only, so tickets with screenshots or attachments need their text extracted first, and the list is unofficial and Jev launched in early access on September 15, 2026. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/route-support-tickets-python/ ## How to try Jev routing rules in the browser Playground with Awesome Jev Repo: [kraayenjon/awesome-jev](https://github.com/kraayenjon/awesome-jev) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README Paste a sample message, add questions, and read typed answers without writing code. **Who it is for:** An operations lead or product person who wants to check whether Jev can make a classification decision before asking engineering to build it. **The problem:** Deciding whether a routing rule is worth building takes time when every test needs a developer to run code. **The outcome:** A short set of tested questions and sample answers that show whether Jev separates the cases you care about. ### Steps 1. Open the Playground at https://console.typesafe.ai/playground. 2. Paste one real example message into the state field. 3. Add a Choice question with three or four options and short descriptions for each. 4. Add a Noul question for a yes or no condition, such as whether the message asks for a refund. 5. Run the question on five to ten examples and note where the answers look wrong. ### Prompt to paste into your agent ```text Using the repo kraayenjon/awesome-jev and the TypeSafe Playground, test a routing rule on . Add one Choice question and one Noul question, run each example, and produce a table of message, choice, confidence, and noul value. Verify that the table covers every sample and flag any answer where the choice does not match the message. ``` ### Caveat The Playground is for testing questions by hand, so it does not replace running the API on your full volume of data. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/test-jev-rules-in-playground/ ## How to add Jev decisions to a JavaScript or TypeScript app with Awesome Jev Repo: [kraayenjon/awesome-jev](https://github.com/kraayenjon/awesome-jev) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Call the Jev API from Node.js with typed Choice, Noul, and Score questions. **Who it is for:** A JavaScript or TypeScript developer who already runs a web or backend service and needs a fast decision step inside a request. **The problem:** Writing a decision step that returns valid, typed output each time is hard when the model returns free text. **The outcome:** A TypeScript function that takes a state object and returns typed answers to named questions. ### Steps 1. Run `npm install @typesafe-ai/sdk`. 2. Set TYPESAFE_API_KEY in your environment, then create a TypeSafeClient. 3. Build a questions object with the choice and noul helpers, as in the README example. 4. Call client.systemOne with your state and questions, then read the answers from the result. 5. Wire the returned choice or noul value into your existing branch logic. ### Prompt to paste into your agent ```text Using the repo kraayenjon/awesome-jev and the @typesafe-ai/sdk npm package, add a function in that takes a request body, sends it to client.systemOne with a choice question and a noul question, and returns both answers. Verify that the function compiles with TypeScript and that the returned object contains an answer for each question name. ``` ### Caveat The README's JavaScript example stops before reading results, so check the result object shape in the official docs before relying on field names. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/add-jev-decisions-to-javascript-app/ ## How to give a coding agent Jev skills for decision steps with Awesome Jev Repo: [kraayenjon/awesome-jev](https://github.com/kraayenjon/awesome-jev) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README Install the TypeSafe agent skill so Claude Code or Codex can build Jev calls. **Who it is for:** A developer using Claude Code or Codex who wants the agent to add classification or gating steps to a codebase without hand-writing each API call. **The problem:** Coding agents often write text-generation calls for decisions that should return typed values, which then need fragile parsing. **The outcome:** An agent that adds Jev-based decision steps to your code and explains which question types it chose. ### Steps 1. Read the agent skill page at https://docs.typesafe.ai/agent-skill. 2. Open https://github.com/typesafe-ai/skills and install the skill in your agent by following that repo's instructions. 3. Set TYPESAFE_API_KEY in the environment the agent runs in. 4. Ask the agent to add one Choice or Noul decision step to a named function in your project. 5. Review the diff, run your tests, and check that the threshold logic matches your intent. ### Prompt to paste into your agent ```text Using the repo kraayenjon/awesome-jev and the TypeSafe agent skill from typesafe-ai/skills, add one Jev Choice decision step to in this project. The step should return a typed label that the existing code branches on. Verify that the project's test suite passes after the change and that the new call reads the API key from TYPESAFE_API_KEY. ``` ### Caveat The skill's install steps live in the typesafe-ai/skills repo, which this list does not reproduce, so follow that source for current instructions. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/kraayenjon-awesome-jev/give-coding-agent-jev-skill/