Ask Eevo via curl (Native API)

Get an API key, send a decision with its consequences, read each option’s expected value, and try Direct mode.

[!NOTE] You will ask Eevo whether to ship a fix now or wait 40 minutes for the logs that would confirm it, for the customers who see a wrong invoice total until it is fixed. You list what each option leads to, and Eevo sizes those consequences against each other. This is Default mode, the one Eevo recommends for supporting agentic and human judgement. At the end you send the same kind of decision in Direct mode. You need a terminal with curl.

1. Get an API key

Sign in to the Eevo console and create a key on API keys. It is shown once: copy it into an environment variable.

export EEVO_API_KEY=eevo_sk_…

Your first account starts with 5 Default mode decisions, each revisable up to 10 times, and 50 Direct mode decisions, free.

2. Send the decision

A decision has a context, two to eight options and the stakeholders whose interests count. Under each option, list its consequences: one concrete effect each that you expect, and whether it is a benefit or a harm. A request without mode runs in Default mode.

curl https://eevo.io/api/v1/decisions \
  -H "Authorization: Bearer $EEVO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "eevo-1",
  "context": "Acme Corp found a data-export bug that shows some customers a wrong invoice total. A fix is ready; a log dump that would confirm it arrives in 40 minutes.",
  "options": [
    {
      "id": "A",
      "description": "Ship the fix now.",
      "consequences": [
        {
          "id": "a1",
          "label": "Wrong totals stop now",
          "description": "About 200 customers stop seeing a wrong invoice total 40 minutes sooner.",
          "valence": "benefit"
        },
        {
          "id": "a2",
          "label": "Invoices to recheck",
          "description": "Without the log dump to confirm the fix, about 20 customers get an email asking them to recheck one invoice.",
          "valence": "harm"
        }
      ]
    },
    {
      "id": "B",
      "description": "Wait for the log dump, then ship.",
      "consequences": [
        {
          "id": "b1",
          "label": "40 more minutes of wrong totals",
          "description": "About 200 customers keep seeing a wrong invoice total for 40 more minutes.",
          "valence": "harm"
        }
      ]
    }
  ],
  "stakeholders": [
    {
      "id": "customers",
      "description": "The affected customers",
      "values": "Correct invoices, no surprises."
    }
  ]
}'

Write the consequences against one baseline, as here: shipping now stops the wrong totals sooner, and waiting keeps them for 40 minutes. Why double counts are discounted explains why that matters.

3. Read the answer

For each stakeholder, the answer has every option's expected value (ev) with a p10–p90 range, and the probability that each option is the best (p_best):

"ev": {
  "A": { "mean": 0.52, "p10": 0.38, "p90": 0.63 },
  "B": { "mean": -0.45, "p10": -0.58, "p90": -0.33 }
},
"p_best": { "A": 0.99, "B": 0.01 }

For the customers, shipping now (A) has the higher expected value. The values are on the stakeholder's own scale, where the largest consequence is 1: compare options within a stakeholder, never across stakeholders. With several stakeholders, Eevo answers for each one side by side and never merges them into one verdict.

links.view_url opens the result in the console: each option's consequences on one axis, and the expected values. links.png_url is the same as an image for a chat message. Add ?stakeholder=<id> for another stakeholder than the first, and format=x for a 1600×900 card sized for social posts, with theme=dark for its dark version.

Keep the id. It finds the decision in the console's decision log, and you need it for the next step.

4. Report what you did

Once you know what was done, tell Eevo. It records, per stakeholder, whether its best option was followed.

curl https://eevo.io/api/v1/decisions/dec_7b1c0e4f9a2d4c6b8e3f5a1d2c4b6e8f/chosen \
  -H "Authorization: Bearer $EEVO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "chosen_option": "A",
  "decided_by": "human"
}'

Open Decisions in the console to see the decision and the option you chose.

5. Get a quick answer in Direct mode

[!NOTE] Above, you listed each option’s consequences. In Direct mode Eevo works them out itself from the context and answers with the best option and a probability for each, in one model call per stakeholder. It is for high-volume decisions and custom tasks, where a fast answer per stakeholder matters more than seeing the reasons.

Same request shape, with "mode": "direct" and no consequences under the options. This one asks about a failed nightly job, for two stakeholders: the customers who read the dashboards and the data team.

curl https://eevo.io/api/v1/decisions \
  -H "Authorization: Bearer $EEVO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "eevo-1",
  "mode": "direct",
  "context": "Our nightly ETL failed at 02:10. A fix is ready, but the log dump that would confirm it arrives at 03:00. Downstream dashboards refresh at 06:00.",
  "options": [
    {
      "id": "A",
      "description": "Ship the fix now"
    },
    {
      "id": "B",
      "description": "Wait for the log dump"
    }
  ],
  "stakeholders": [
    {
      "id": "customers",
      "description": "The customers who read the dashboards",
      "values": "Correct numbers on time"
    },
    {
      "id": "data-team",
      "description": "The data team",
      "values": "Reliability first; no repeated incidents"
    }
  ]
}'

The answer:

{
  "id": "dec_4e9a1c7b2d8f4a3e9c6b0d1f7a2e5c84",
  "model": "eevo-1",
  "results": [
    {
      "stakeholder": {
        "id": "customers",
        "description": "The customers who read the dashboards"
      },
      "best": "A",
      "confidence": 0.74,
      "probabilities": {
        "A": 0.74,
        "B": 0.26
      }
    },
    {
      "stakeholder": {
        "id": "data-team",
        "description": "The data team"
      },
      "best": "B",
      "confidence": 0.63,
      "probabilities": {
        "A": 0.37,
        "B": 0.63
      }
    }
  ],
  "agreement": {
    "all_same": false,
    "best_by_stakeholder": {
      "customers": "A",
      "data-team": "B"
    }
  },
  "usage": {
    "input_tokens": 2381,
    "output_tokens": 0
  },
  "cost_usd": 0.00595
}

Eevo ran once per stakeholder. For the customers, shipping now (A) is best, with probability 0.74. For the data team, waiting (B) is best, with 0.63. agreement.all_same is false: the stakeholders disagree, and Eevo leaves that trade-off to you.

Next