Point your agent at it.

Read https://megachadcua.com/llms.txt and add megachad to this project: <what it should decide, e.g. route each support ticket to Billing, Bug or Other>. Use the API key from the MEGACHAD_API_KEY env var; if it's unset, ask me. Never hardcode or commit the key. Call the HTTP API directly unless the project already uses a megachad SDK.

mkdir -p ~/.claude/skills/megachad && curl -fsSL https://megachadcua.com/agents/megachad/SKILL.md -o ~/.claude/skills/megachad/SKILL.md

claude mcp add megachad -e MEGACHAD_API_KEY=$MEGACHAD_API_KEY -- npx -y megachad-mcp

curl -fsSL https://megachadcua.com/agents/AGENTS.md >> AGENTS.md

## megachad decision API

megachad (https://megachadcua.com) answers a schema of questions about text or speech: each field is one of N options, yes/no, or a scale, and comes back with a calibrated confidence. The model decides in ~17 ms. Full reference: https://megachadcua.com/llms-full.txt.

Key:

- The key is in the env var `MEGACHAD_API_KEY` (`mcc_live_...`). Read it at runtime. Never hardcode, commit, print or log it; never send it to a browser.
- If it's unset, stop and ask the user to get one at https://megachadcua.com/login?signup=1 and set it. Don't sign up for them.
- Keep `.env` in `.gitignore`.

Call (raw HTTP is the default; no SDK needed):

```bash
curl https://api.megachadcua.com/v1/decide \
  -H "Authorization: Bearer $MEGACHAD_API_KEY" \
  -H "content-type: application/json" \
  -d '{"schema": {"action": {"question": "What should the agent do?", "options": ["Click Sign in", "Scroll down", "Nothing yet"]}}, "text": "take me to sign in"}'
```

Response (200):

```json
{"session_id": "dcd_C8o5VFBm5AH65L7e", "fields": {"action": {"value": "Click Sign in", "confidence": 0.98, "probabilities": {"Click Sign in": 0.98, "Scroll down": 0.012, "Nothing yet": 0.008}}}, "transcript": "take me to sign in", "latency_ms": 112, "model_ms": 15.2, "usage": {"credits": 1}}
```

- Field kinds: `{"question": "...", "options": ["A", "B", "Nothing yet"]}`, `{"question": "...", "type": "yes_no"}`, `{"question": "...", "scale": ["low", "medium", "high"]}`.
- `text`: a string, or `[{"speaker": "user" | "other", "text": "..."}]`.
- Up to 50 texts per call: `POST /v1/decide/batch` with `{"schema", "items": [{"id", "text"}]}`. Live voice: `WS /v1/listen`.
- Schema: one question per field; short, distinct options with a way out ("Nothing yet", "Not sure"); option labels as the user sees them. The value comes back verbatim.
- Act only at or above a confidence threshold (start at 0.85); below it, confirm with the user. `confidence` can be `null`: treat as unsure. A human confirms destructive, financial or medical actions. Medical: it routes and structures for clinicians; it doesn't diagnose.
- Errors: `{"error": {"code", "message"}}`. Retry 409, 429 and 5xx with backoff and `Retry-After`, sending the same `Idempotency-Key` (a UUID per request) so a retry is never billed twice. Don't retry 400, 401, 402, 403, 413, 422; 402 and 403 need the user.
- SDKs: `pip install megachad` or `npm i megachad` (0.1.0: `decide`, `listen`); MCP: `npx -y megachad-mcp`. SDK batch and retry helpers are coming in 0.1.1; don't use them yet.
- Tests mock the HTTP call; real calls cost 10¢ per 1,000.

mkdir -p .cursor/rules && curl -fsSL https://megachadcua.com/agents/megachad.mdc -o .cursor/rules/megachad.mdc

{
  "mcpServers": {
    "megachad": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "megachad-mcp"
      ],
      "env": {
        "MEGACHAD_API_KEY": "${env:MEGACHAD_API_KEY}"
      }
    }
  }
}

curl https://api.megachadcua.com/v1/decide \
  -H "Authorization: Bearer $MEGACHAD_API_KEY" \
  -H "content-type: application/json" \
  -d '{"schema": {"action": {"question": "Which on-screen control does the user want the agent to use?", "options": ["Sign in", "Pricing", "Docs", "Nothing yet", "Not sure"]}}, "text": "take me to sign in"}'

{
  "name": "megachad_decide",
  "description": "Make fast, calibrated decisions about a piece of text with megachad (the model decides in ~17 ms). Give a schema of fields; each field is one question answered by picking one of its options, yes/no, or a level on a scale. Returns each field's value and a confidence from 0 to 1. Use it to map what a person said to one of a fixed set of actions, intents, queues or labels. Below about 0.85 confidence, ask the person instead of acting.",
  "input_schema": {
    "type": "object",
    "properties": {
      "schema": {
        "type": "object",
        "description": "Field name to field. One decision per field. Each field has a question and exactly one of: options (pick one), type yes_no, or scale (ordered levels, low to high).",
        "additionalProperties": {
          "type": "object",
          "properties": {
            "question": {
              "type": "string",
              "description": "The question the model answers about the text."
            },
            "options": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "description": "Pick one of these. The answer is the option text, verbatim. Include a way out like \"Nothing yet\" or \"Not sure\"."
            },
            "type": {
              "type": "string",
              "enum": [
                "yes_no"
              ],
              "description": "yes_no: the answer is true or false."
            },
            "scale": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "description": "Ordered levels, low to high. The answer is a level, plus its 0-based index."
            }
          }
        }
      },
      "text": {
        "type": "string",
        "description": "What the person said or wrote: an utterance, a transcript, a ticket."
      }
    },
    "required": [
      "schema",
      "text"
    ]
  }
}

{
  "type": "function",
  "function": {
    "name": "megachad_decide",
    "description": "Make fast, calibrated decisions about a piece of text with megachad (the model decides in ~17 ms). Give a schema of fields; each field is one question answered by picking one of its options, yes/no, or a level on a scale. Returns each field's value and a confidence from 0 to 1. Use it to map what a person said to one of a fixed set of actions, intents, queues or labels. Below about 0.85 confidence, ask the person instead of acting.",
    "parameters": {
      "type": "object",
      "properties": {
        "schema": {
          "type": "object",
          "description": "Field name to field. One decision per field. Each field has a question and exactly one of: options (pick one), type yes_no, or scale (ordered levels, low to high).",
          "additionalProperties": {
            "type": "object",
            "properties": {
              "question": {
                "type": "string",
                "description": "The question the model answers about the text."
              },
              "options": {
                "type": "array",
                "items": {
                  "type": "string"
                },
                "description": "Pick one of these. The answer is the option text, verbatim. Include a way out like \"Nothing yet\" or \"Not sure\"."
              },
              "type": {
                "type": "string",
                "enum": [
                  "yes_no"
                ],
                "description": "yes_no: the answer is true or false."
              },
              "scale": {
                "type": "array",
                "items": {
                  "type": "string"
                },
                "description": "Ordered levels, low to high. The answer is a level, plus its 0-based index."
              }
            }
          }
        },
        "text": {
          "type": "string",
          "description": "What the person said or wrote: an utterance, a transcript, a ticket."
        }
      },
      "required": [
        "schema",
        "text"
      ]
    }
  }
}