# Eevo > Eevo is a strong general-purpose SystemOne model ([decision model](https://simonwillison.net/2026/Sep/21/jev/)), especially good in tasks that involve understanding and comparing consequences, estimating expected value, systems thinking and stakes. Like other SystemOne models, Eevo can be used via the SystemOne-compatible API, which answers questions (choice, score, yes/no) about any state, for tasks such as model routing, ranking and content filtering. Eevo also has a native API for consequence-based decisions, which you can also use via MCP: it takes a decision with its options, their consequences and the stakeholders, and answers with each option’s expected value per stakeholder. ## Tutorials - [Calling Eevo from code](https://eevo.io/api/docs/tutorials/questions.md): Call Eevo from code (Python, TypeScript or cURL). - [Connecting your agent to Eevo](https://eevo.io/api/docs/tutorials/agent.md): Add Eevo’s MCP server to Claude, ChatGPT, Claude Code, Codex or another MCP client, so your agent asks Eevo when a choice affects people. ## How-to guides - [Model routing with Eevo](https://eevo.io/api/docs/how-to/model-routing.md): Eevo analyzes each task’s attributes (difficulty, category, stakes, tools, sensitivity), and your routing policy picks the model. - [Helping agents consider consequences](https://eevo.io/api/docs/how-to/consequential-decisions.md): Eevo’s Consequence-based Decision API helps AI agents and humans make better decisions based on their consequences. - [Detecting hazards from photos with Eevo](https://eevo.io/api/docs/how-to/site-safety-photos.md): Eevo answers questions about site photos, finds the hazard in the photo, and weighs what to do about it. - [Ranking search results with Eevo](https://eevo.io/api/docs/how-to/ranking.md): Eevo scores how well each search result answers the query, and you sort the results by the scores. - [Content filtering with Eevo](https://eevo.io/api/docs/how-to/content-filtering.md): Eevo says whether a text breaks your policy, so you publish it, hold it for a person, or reject it. - [Injection safeguards with Eevo](https://eevo.io/api/docs/how-to/injection-safeguards.md): Eevo checks untrusted text for instructions aimed at your agent before the agent reads it, so you can stop or flag it. - [PII detection with Eevo](https://eevo.io/api/docs/how-to/pii-detection.md): Eevo says whether a text contains personally identifiable information, and which kinds, so your safeguards can act before it leaks. - [Investment thesis match screening with Eevo](https://eevo.io/api/docs/how-to/thesis-screening.md): Eevo evaluates whether each company in a list matches an investment thesis, from natural language descriptions of the thesis and the company. - [Process photos in insurance claims with Eevo](https://eevo.io/api/docs/how-to/claim-photos.md): Eevo reads damage photos for what is likely to happen next, and weighs how to handle the claim. - [Writing requests for Eevo](https://eevo.io/api/docs/how-to/writing-requests.md): How to write the questions for the SystemOne-compatible API, and the options and consequences for the Consequence-based Decision API. ## Reference - [Authentication](https://eevo.io/api/docs/reference/authentication.md): Requests carry an API key in the Authorization header; MCP clients can sign in with OAuth instead. - [API reference](https://eevo.io/api/docs/reference/api.md): Find every Eevo API endpoint, grouped by what it answers, with its fields. - [SystemOne decision](https://eevo.io/api/docs/reference/api-systemone.md): Ask questions about any state (choice, score or yes/no) and get each answer with probabilities. Images and box questions are in preview. - [Consequence-based decision](https://eevo.io/api/docs/reference/api-cb-decision.md): Weigh consequences, model their overlap and compute expected value for one or more stakeholders. - [Get a decision](https://eevo.io/api/docs/reference/api-get-decision.md): Fetch a decision again, with its chosen decision, revisions and assessments. - [Share a decision](https://eevo.io/api/docs/reference/api-sharing.md): Choose who sees a decision: everyone in your organization, or only you. - [Report what was done](https://eevo.io/api/docs/reference/api-chosen.md): Report the option that was carried out, for the whole decision. - [Publish a decision page](https://eevo.io/api/docs/reference/api-publication.md): Give a decision a public page anyone with the link can read, or take it down. - [Record whether you agree](https://eevo.io/api/docs/reference/api-assessments.md): Record your verdict on the answer, per stakeholder. - [List payment receipts](https://eevo.io/api/docs/reference/api-receipts.md): List your organization’s payments with their Stripe receipts, for bookkeeping. - [Models, pricing and limits](https://eevo.io/api/docs/reference/models-pricing.md): Eevo-1’s model ids, the price of each API, and the limits a request must fit. - [Data controls](https://eevo.io/api/docs/reference/data-controls.md): Choose how long each key’s requests are kept, whether Eevo can learn from them, and where they run. - [Errors](https://eevo.io/api/docs/reference/errors.md): Every error has a status, a type your code can branch on, and a message to show. ## Appendix - [How Eevo’s Consequence-based Decision API works](https://eevo.io/api/docs/explanation/how-eevo-decides.md): Eevo weighs each option by what it leads to for each stakeholder, and counts an effect listed twice only once. - [Probability, confidence and expected value](https://eevo.io/api/docs/explanation/probability-and-ev.md): A probability says how likely an option is to be the best one, and an expected value says how good it is on the stakeholder’s own scale. ## API - [OpenAPI 3.1](https://eevo.io/api/openapi.json): every endpoint and field