[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_113":114},"routes/modules.$creator.$moduleId",{"_14":15,"_16":17,"_18":19,"_20":21},"routes/modules.$creator.$moduleId.endpoints",{"_11":12,"_13":-5},"error","Module not found","tools","module",{"_20":22,"_23":24,"_25":26,"_27":28,"_29":30,"_31":32,"_33":34,"_35":26,"_36":37,"_38":-5,"_39":-5,"_40":41,"_42":43,"_44":45,"_46":47,"_48":49,"_50":51,"_52":53,"_54":55,"_56":57,"_58":-5,"_59":51,"_60":45,"_61":62,"_63":64,"_65":66,"_67":68,"_69":70,"_71":72,"_73":74,"_75":76,"_77":78,"_79":80,"_81":-5,"_82":-5,"_83":-5,"_84":85,"_86":-5,"_87":45,"_88":89,"_90":91,"_92":93,"_94":95,"_96":97,"_98":99},"apiUrl","https://odel.app/api","creator","turingcorp","moduleId","decider","@turingcorp/decider","displayName","TuringCorp","tagline","A decision model for the calls that don't have a right answer. The better option, with a reason.","iconKey","external/ee5a03fb1dc98f7dc1840083d426cf80b224a0dbb7c8e04c82f9452bcc5a10d4.jpg","category","remote","status","published","visibility","public","metaDescription","deploymentType","external","workerName","bindingName","mcpConfig","{\"url\":\"https://mcp.turingcorp.net/mcp\",\"transport\":\"streamable-http\",\"authType\":\"bearer\"}","pricingModel","free","priceCredits",0,"creatorUserId","system:external-import","createdAt",1790145536000,"updatedAt",1790466697000,"sourceRegistry","official-mcp-registry","sourceUrl","https://registry.modelcontextprotocol.io/v0/servers?search=net.turingcorp.mcp%2Fdecider","sourceVerified",true,"sourceMissingSince","importedAt","upstreamStars","upstreamUpdatedAt",1790387724000,"detectedTransport","streamable-http","detectedAuthType","none","probeStatus","ok","lastProbedAt",1790453317000,"githubRepoUrl","https://github.com/TuringCorp-net/MCP","githubRepoFullName","TuringCorp-net/MCP","releaseId","ext-turingcorp-net-mcp","version","1.0.3","descriptionMarkdown","# TuringCorp — MCP server\n\nA **decision model** for the calls that don't have a right answer. Your agent has two defensible options and has to pick one. Send both: a **panel of models** judges them together and returns the better one, **how far apart it judged them** (a calibrated confidence), and why.\n\n- **Endpoint (canonical):** `https://mcp.turingcorp.net/mcp`\n- **Aliases:** `https://mcp.turingcorp.net/` and `https://mcp.turingcorp.net/mcp/`\n- **Health:** `https://mcp.turingcorp.net/healthz`\n- **Transport:** Streamable HTTP, **stateless** (no session, no Durable Object). `GET /mcp` → 405 is normal.\n- **Protocol:** MCP `2026-07-28` (modern) **and** `2025-06-18` (legacy) — one route serves both.\n- **Discovery methods** (`server/discover`, `tools/list`) need **no credentials** — directory scanners can read the full tool list unauthenticated, by design.\n\n## The one tool: `decide`\n\n| | |\n|---|---|\n| **In** | `task` (string), `optionA` (string), `optionB` (string) — all required |\n| **Out** | `job_id` (string), `betterOption` (`\"option_A\"` \\| `\"option_B\"`), `confidence` (percentage string, e.g. `\"76.7%\"`), `reason` (string) |\n| **Annotations** | `readOnlyHint: true` · `openWorldHint: false` · `idempotentHint: false` |\n| **Timeout** | Reserve **180–300 seconds** — a decision is a long call. The timeout is a **client/host setting, not a tool parameter**: there is nothing to pass in the call. A 60s default cuts it off before the answer arrives; if that happens, do not call again — retrieve by `job_id`. |\n| **Long calls** | A client that declares the `io.modelcontextprotocol/tasks` extension gets a task handle back instead of holding one connection open for minutes, and polls `tasks/get` — see [Retrieving a result](#retrieving-a-result). Clients that do not declare it see no change at all. |\n\n**A successful call returns the decision inline.** `job_id`, `betterOption`, `confidence` and `reason` all arrive\nin the *same* tool result — there is nothing to poll and nothing to fetch afterwards. The `job_id` is only for the\ncase where the call never came back (timeout, dropped connection, client gave up waiting).\n\n`confidence` is **this service's own judgement of how far apart the two options were** — a reference for your decision-making, not an instruction, not a result, and not a prediction of how the choice turns out. As a rough reading: ≥90% clearly apart · 80–90% apart, less clearly · 70–80% a closer call · <70% close to evenly matched. These ranges are descriptive only. **Choose your own threshold for your own use case; for high-stakes or irreversible decisions apply your own review policy.** Observed accuracy for each range, and how it was measured, is published at — that page is the authority on the bands and their measured accuracy.\n\n⚠️ `idempotentHint: false` is an honest declaration: a retried call is a new call. **Record the `job_id` the result returns** — a decision you have already paid for can be collected for **7 days** with the same credential (`GET https://api.turingcorp.net/v1/jobs?job_id=`; without the id, `GET https://api.turingcorp.net/v1/jobs` lists the job ids for that credential). Retrieve instead of retrying.\n\n## When not to use it\n\n- **More than two options.** It compares exactly A and B — there is no third slot, and it will not rank a list.\n- **Anything you can compute or verify.** A spec, a test, a price, a document: if something objective decides it, use that. This is for choices where no objective rule does.\n- **Factual questions.** It picks between two candidates; it does not look anything up.\n- **Speed-critical paths.** A decision is a long call and a paid one. Do not put it behind a request that has to answer in seconds.\n- **High-stakes irreversible calls without review.** Route on the confidence and keep your own review policy.\n\n## How to fill the three arguments\n\n- `task` — state the decision **neutrally**, without leaning toward either side: *\"Which email do I send?\"*, not *\"Should I send the honest one?\"*\n- `optionA` / `optionB` — one **concrete** option each, plus the case for it. Plain text or Markdown, any length; keep the two sides roughly comparable so the comparison is fair.\n- **One option = one plan.** Do not bundle alternatives into a single side (\"go indoors *or* postpone\"): it compares the two slots, it does not split one of them for you.\n\n```json\n{\n \"task\": \"Which version of the delivery-slip email do I send to a client we want to keep?\",\n \"optionA\": \"Short and direct: the integration took longer than planned, delivery moves to the 24th, everything else is unchanged.\",\n \"optionB\": \"Warmer and longer: thank them for the kickoff, explain that dependencies took more time, offer to walk through the details.\"\n}\n```\n\nMore patterns — choosing the version you send, thresholding on the confidence, and writing the two options so\nthe comparison is fair: **[docs/COOKBOOK.md](https://github.com/TuringCorp-net/MCP/blob/main/docs/COOKBOOK.md)**.\n\n## Retrieving a result\n\nEvery call returns a `job_id`. If the call times out or the connection drops, **do not call again** — retrieve it:\n\n- **With the id** — `GET https://api.turingcorp.net/v1/jobs?job_id=` with the same Agent Pass. A client that speaks the `io.modelcontextprotocol/tasks` extension can use `tasks/get` with `{\"taskId\":\"\"}` on the MCP endpoint instead.\n- **Without it** — `GET https://api.turingcorp.net/v1/jobs` lists the job ids that credential created in the last 7 days; then fetch one as above. The list carries only the job id, the product and a timestamp — fetch a job by id to see what it was.\n- **Nothing yet** — an empty list is **not** an error, and there is **no** `job_id=0` placeholder: `{\"object\":\"list\",\"window_seconds\":604800,\"data\":[]}`. Asking for `0` returns `404 No such job.`\n\nRetrieval returns the job's status and, once it succeeded, the same body the call itself would have returned. A job that is not yours, or older than **7 days**, is reported as unavailable. `GET https://api.turingcorp.net/v1/account` returns `{\"account_id\":\"…\"}`.\n\n## See it decide first\n\nThere is no free tier, so we publish the evidence instead: **27 real decisions, recorded verbatim** — the question, both options, which one was preferred, the confidence reported, and the reason. Nine domains, three cases each: Writing · Tech · Business · Research · Career · Money · People · Travel · Everyday.\n\n→ **** — dataset: [`examples.json`](https://github.com/TuringCorp-net/poe-demo-public/blob/main/examples.json)\n\nRead the confidence column first. On ordinary, closely matched questions it reports **70–90%, not 99%** — and that is the point of a calibrated number rather than a decorative one: a high value means the comparison was decisive and the pick can be acted on, a low value means the two options really are close and the choice stays yours. In this set the range is **27.3%–88.3%, median 74.0%, none above 90%**, because these are everyday close calls, not easy ones. What the bands deliver when they *are* decisive is published at .\n\n## Authentication\n\nSend an **Agent Pass**: `Authorization: Bearer `.\n\nIssued at **** — self-service signup (email verification) → top up → get a pass. A pass is valid for **7 days** and can be re-rolled. If a call is refused as an invalid credential, sign in there again and re-roll.\n\nErrors fall into **two classes**, and they arrive in different places:\n\n> Credential problems are rejected before the call (HTTP 401 + `WWW-Authenticate`). Business failures come back as a tool result with `isError: true` plus a JSON block `{error, http_status, action_url, message}` — `http_status` is the upstream status; the tool call itself is HTTP 200.\n\n| Class | What comes back | Examples |\n|---|---|---|\n| **Credential** | HTTP `401` + `WWW-Authenticate` — no tool result is produced | no credential · expired or invalid pass (`invalid_credential`, with a re-login URL in the message) |\n| **Business** | HTTP `200`, tool result `isError: true`, plus a JSON block | insufficient balance (`http_status` `402`, `action_url` pointing at top-up) · over quota (`http_status` `429`) |\n\nBranch on the **HTTP status first**: a `401` will not succeed on retry without a new credential, whereas a business failure was a call that ran and was declined — read the block for the reason and the next step. The detailed contract is under *Errors are machine-readable* below.\n\nEntry rate limit: **120 requests / 60 s / client IP**. This is flood damping, not a quota — the real per-key quota is enforced separately. The counter is approximate: it is maintained per edge location and is eventually consistent, so a brief overshoot is possible.\n\n## Two ways in — and they are not the same\n\n- **Through an MCP client or host** (Claude Code, Cursor, VS Code, Codex, TRAE, Coze…). The host holds the credential and attaches it for you: **you do not set an `Authorization` header yourself, and in most hosts you cannot**. The timeout is a host setting, not something you pass in the call.\n- **Directly against the REST API** (). Here you *do* send `Authorization: Bearer ` yourself, and you can retrieve a job by id.\n\n⚠️ The part that catches people out: **being able to call the tool through a host does not mean you can reach the\nREST API.** The host may never hand you the underlying Agent Pass, so a job-retrieval call you make on your own can\ncome back `401`. If your host declares the Tasks extension, retrieve through the tool surface instead (see\n[Retrieving a result](#retrieving-a-result)); if it does not, retrieval may simply not be reachable from where you are.\n\n## Quickstart\n\nYou need an **Agent Pass** to call the tool. Discovery (`tools/list`) works without one — calling `decide` does not.\n\n🔑 **Where the pass lives matters.** If your client can keep it out of the file, do that — VS Code's\n`${input:…}`, Codex's `bearer_token_env_var`. A pass written in plain text inside `mcp.json` or any client config\nis readable by **every agent and process that can read that file**, and assistants do read their own config — one\ncan print your pass straight into its output. Treat a client config as public within your machine.\n\n### 1. Get an Agent Pass\n\nSelf-service at ****: sign up with an email address, verify it, add credit. The pass is shown once; it is valid for **7 days** and can be re-rolled at any time. Agents can do the same thing over the API — see that site's `/llms.txt`.\n\n### 2. Point your client at the endpoint\n\n**Any client that reads a JSON config — Cursor, Cline, Windsurf, Claude Desktop:**\n\n```json\n{\n \"mcpServers\": {\n \"TuringCorp\": {\n \"type\": \"http\",\n \"url\": \"https://mcp.turingcorp.net/mcp\",\n \"headers\": { \"Authorization\": \"Bearer \" }\n }\n }\n}\n```\n\nKeep the `\"type\": \"http\"` line: a client that reads a `url` entry with no `type` treats it as a local stdio\nserver and skips it.\n\n**A client that only speaks stdio** needs a bridge, and the credential must go in through `--header` —\n`mcp-remote` does not read an `AUTHORIZATION` environment variable, so a config that only sets one connects,\nlists tools, and then fails on the first call with 401. The missing space after `Authorization:` is deliberate:\nsome clients mangle spaces inside `args`.\n\n```json\n{\n \"mcpServers\": {\n \"TuringCorp\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"mcp-remote\", \"https://mcp.turingcorp.net/mcp\",\n \"--header\", \"Authorization:${AGENT_PASS}\"],\n \"env\": { \"AGENT_PASS\": \"Bearer \" }\n }\n }\n}\n```\n\n> ⚠️ A bare `npx -y mcp-remote@latest https://mcp.turingcorp.net/mcp` with no credential **will connect and list tools, then fail on the first call** with 401. That is expected — discovery is open, execution is not.\n\n**Claude Code:**\n\n```bash\nclaude mcp add --transport http TuringCorp https://mcp.turingcorp.net/mcp \\\n --header \"Authorization: Bearer \"\n```\n\n### 3. Sanity-check it yourself\n\n**Discovery — no credentials needed:**\n\n```bash\ncurl -s -X POST https://mcp.turingcorp.net/mcp \\\n -H 'content-type: application/json' \\\n -H 'accept: application/json, text/event-stream' \\\n -H 'mcp-protocol-version: 2026-07-28' \\\n -H 'mcp-method: tools/list' \\\n --data '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"tools/list\",\"params\":{\"_meta\":{\"io.modelcontextprotocol/protocolVersion\":\"2026-07-28\",\"io.modelcontextprotocol/clientCapabilities\":{}}}}'\n```\n\n**Calling the tool** additionally needs `mcp-name: decide` and the credential:\n\n```bash\ncurl -s -X POST https://mcp.turingcorp.net/mcp \\\n -H 'content-type: application/json' \\\n -H 'accept: application/json, text/event-stream' \\\n -H 'mcp-protocol-version: 2026-07-28' \\\n -H 'mcp-method: tools/call' \\\n -H 'mcp-name: decide' \\\n -H \"authorization: Bearer $AGENT_PASS\" \\\n --data '{\"jsonrpc\":\"2.0\",\"id\":2,\"method\":\"tools/call\",\"params\":{\"name\":\"decide\",\"arguments\":{\"task\":\"Pick a launch date\",\"optionA\":\"Ship now\",\"optionB\":\"Wait two weeks\"},\"_meta\":{\"io.modelcontextprotocol/protocolVersion\":\"2026-07-28\",\"io.modelcontextprotocol/clientCapabilities\":{}}}}'\n```\n\n### Versions\n\nTwo version numbers exist and they are **not** meant to match:\n\n| | meaning |\n|---|---|\n| **Registry `version`** (`server.json`) | the **listing** version - the metadata record (endpoint, auth declaration, icon, links). Advances when the listing changes. |\n| **`serverInfo.version`** (from the live endpoint) | the **runtime** version of the deployed server. Advances when behavior changes. |\n\nA listing-only change (for example a corrected link) advances the registry version without\ntouching the runtime, so the two can legitimately differ.\n\n### Protocol details worth knowing\n\n- **`Accept` must contain both** `application/json` **and** `text/event-stream`, or you get 406.\n- **Protocol versions:** one route serves both generations.\n - **Modern `2026-07-28`** — no `initialize` handshake; use `server/discover` and `_meta`.\n - **Legacy** — standard `initialize` handshake. Measured negotiation: a client asking for `2025-11-25`, `2025-06-18`, or `2025-03-26` is answered with **exactly the version it asked for**; a modern `2026-07-28` `initialize` (which is not part of that generation) is answered with `2025-11-25`.\n - Server capability discovery via `tools/list` and `server/discover` needs no credentials.\n- **Errors are machine-readable.** There are two distinct kinds, and they arrive differently — so branch on the right thing:\n\n| What failed | How you see it | What to branch on |\n|---|---|---|\n| **Credential** missing or invalid | **HTTP `401`** with a `WWW-Authenticate` challenge (carrying `error=\"invalid_token\"` and `resource_metadata`); the body is a standard JSON-RPC error with `error.code = -32001`. No tool result is produced. | the **HTTP status code** |\n| **Business** failure — insufficient balance, quota, upstream error | a normal tool result (`isError: true`) whose **second `content` block** is a single JSON object you can `JSON.parse`, e.g. `{\"error\":\"insufficient_balance\",\"http_status\":\"402\",\"action_url\":\"…\",\"message\":\"…\"}` | the **`error` field** inside that block |\n\n In short: credential problems are rejected before the tool runs, so they never appear as a tool result; everything else that fails after that does, with a machine-readable block attached.\n- **Authorization discovery:** `401` responses carry a `resource_metadata` pointing at `/.well-known/oauth-protected-resource`. Note that this server uses a **static bearer credential, not OAuth** — that document says so explicitly rather than sending you into an OAuth flow that does not exist.\n\n## Not for browser clients\n\nThis endpoint deliberately does **not** serve browser-based (cross-origin) MCP clients: requests carrying a browser `Origin` header other than localhost are rejected with 403, and no CORS headers are returned. Browser tooling such as the MCP Inspector works by proxying through a **local** process, which sends no `Origin` — that path works normally. If you need a browser page to reach this server, run a proxy you control rather than calling it cross-origin.\n\n## What this is not\n\n- **One tool only.** `decide` is the entire surface. There is no way to reach other tiers through this endpoint.\n- **No SLA.** No availability commitment is offered, and none should be inferred.\n- **Not an autopilot.** Decider is a component you call. How you gate on it — thresholds, human review, retries — is your policy and stays yours.\n- **No idempotency key — record the job id instead.** A retried call is a new call; collect the result with the `job_id` instead.\n\n## Use it from a coding agent: the Agent Skill\n\nThere is an **[Agent Skill](https://github.com/TuringCorp-net/MCP/blob/main/skills/decider/SKILL.md)** for coding agents (Claude Code, Codex, and anything that\nreads the same skill convention). It teaches an agent *when* a two-option judgement is the right tool, how to\nwrite the two options so the comparison is fair, and how to read the confidence without over-reading it.\n\nInstall it by pointing your agent at this repository, or copy `skills/decider/` into your skills directory and\nadd the MCP server to your client (see [Quickstart](#quickstart)).\n\n## About this repository\n\nThis repository carries the **discovery metadata** for the server above — what directory services and clients read to find and describe it.\n\n- [`server.json`](https://github.com/TuringCorp-net/MCP/blob/main/server.json) — the official MCP Registry manifest\n- [`docs/DIRECTORIES.md`](https://github.com/TuringCorp-net/MCP/blob/main/docs/DIRECTORIES.md) — where this server is listed, and what each directory requires\n- [`docs/COOKBOOK.md`](https://github.com/TuringCorp-net/MCP/blob/main/docs/COOKBOOK.md) — worked patterns for handing `decide` the right kind of call\n- [`skills/decider/`](https://github.com/TuringCorp-net/MCP/blob/main/skills/decider/SKILL.md) — the Agent Skill\n\nThis server is also listed on Smithery: ****\n\nThe server itself is a hosted remote endpoint; nothing here needs to be installed or run.\n\n## License\n\nSee [LICENSE](https://github.com/TuringCorp-net/MCP/blob/main/LICENSE).\n","changelogMarkdown","videoKey","requiredSecretsJson","publishedAt",1790387582000,"priceCents","trialDays","iconType","image","upstream",{"_102":103,"_104":105,"_106":-5,"_107":108,"_109":110,"_111":112},"requiredSecrets",[],"tags",[],"assets",[],"stats",{"_100":45,"_101":45},"ratingAvg","reviewCount","topics",[],"license","MIT","language","registryName","net.turingcorp.mcp/decider","homepage","https://mcp.turingcorp.net/","packages",[],"ENV",{"_115":116,"_117":118,"_119":120,"_121":-7},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272","SENTRY_ENABLE_LOCAL"]