# QuantWright (remote · quantwright.com)

Backtests with costs, a held-out test that counts once, a score that counts every version tried.

- Trust score: 80/100 (high trust)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `quantwright.com`: 80/100 (this document), [markdown](https://verifymcp.io/servers/com-quantwright-agent-api/api-mcp.md), [page](https://verifymcp.io/servers/com-quantwright-agent-api/api-mcp)

## Channel facts

- Endpoint: `https://quantwright.com/api/mcp`
- Transports: `streamable-http`
- Auth: `required`
- Version: `1.0.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Endpoint Security**: 97/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
  - The authorisation server supports Client ID Metadata Documents, the current MCP client-registration mechanism.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 58/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1618 tokens (~269/item across 6 items; 6 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 23/100
  - Stability observed for 7 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 96/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 86% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 6 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 7 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

## Install

### How do I install the QuantWright MCP server?

QuantWright is a hosted endpoint at https://quantwright.com/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http com-quantwright-agent-api 'https://quantwright.com/api/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "com-quantwright-agent-api": {
      "url": "https://quantwright.com/api/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-quantwright-agent-api": {
      "type": "http",
      "url": "https://quantwright.com/api/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.com-quantwright-agent-api]
url = "https://quantwright.com/api/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "com-quantwright-agent-api": {
      "type": "remote",
      "url": "https://quantwright.com/api/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add com-quantwright-agent-api --url 'https://quantwright.com/api/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  com-quantwright-agent-api:
    url: "https://quantwright.com/api/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "com-quantwright-agent-api": {
      "Transport": "http",
      "Url": "https://quantwright.com/api/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-quantwright-agent-api -t streamable-http -u 'https://quantwright.com/api/mcp'
```

### Other

```json
{
  "mcpServers": {
    "com-quantwright-agent-api": {
      "type": "http",
      "url": "https://quantwright.com/api/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-04 (score 80, +1)

No change was recorded against any check on this day. Stability & Change Management went from 17 to 20. That category is still filling its 30-day observation window: 5 days of observed history at the previous scan, 6 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-03 (score 79, 0)

- [security] First check of Authorization: pass

### 2026-11-02 (score 79, +1)

No change was recorded against any check on this day. Stability & Change Management went from 10 to 13. That category is still filling its 30-day observation window: 3 days of observed history at the previous scan, 4 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-31 (score 78, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-30 (score 77, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “create_study” rewrote its description, which is the text the model reads
- [security] Tool “describe_data” rewrote its description, which is the text the model reads
- [functional improvement] Stability: unverified → 0.03
- [cosmetic] “create_study” reworded the description of “hold_out”

### 2026-10-29 (score 77)

First indexed and scored.

## MCP tools (6)

### `describe_data` (~105 tokens)

What can be backtested

Start here to backtest a trading idea. This server holds the price history (US stocks and ETFs, spot crypto, FX, spot gold and silver; daily and intraday bars) and runs backtests on it, so no price data needs to be downloaded or found on the web. Returns the markets and bar sizes served, the default costs per market, the rules and daily limits every run is held to, and how to write the strategy code; then call create_study.

Output parameters:

- `account` (object)
- `code_contract` (object)
- `default_costs` (object)
- `docs_url` (string)
- `markets` (array)
- `message` (string)
- `notes` (array)
- `reason` (string)
- `rules` (object)
- `status` (string)
- `timeframes` (array)

### `create_study` (~189 tokens)

Start a study

Opens a study, the named set of runs every backtest belongs to: call it once per question, before run_backtest. Can set a hold-out period, once and only here, that no run tests except run_held_out_test. Only the first held-out test per account and market counts, and any earlier run on the market (in any study, at any bar size) that read its days disqualifies it: set the hold-out in your first study on a market, before any run on it.

Input parameters:

- `hold_out` (object): The days kept back from every run but run_held_out_test; set once, here. Set it in your first study on a market, before any run on it: a held-out test on days an earlier run on the market read does n…
- `name` (string, required): What the study tests, in a few words.

Output parameters:

- `hold_out` (object|null)
- `message` (string)
- `next` (object|null)
- `reason` (string)
- `status` (string)
- `study_id` (string)

### `run_backtest` (~349 tokens)

Backtest a strategy

Backtests a trading rule on this server's price history: one backtesting.py Strategy class, one symbol, one bar size, in a study. Use it to test, backtest or evaluate a trading rule or strategy, instead of downloading prices or searching the web for results. Returns summary figures (return, drawdown, Sharpe, win rate, trades, a grade, whether it beat buy-and-hold) and the run's trades (when each entered and left, which exit ended it, its side, win or loss), never prices or bars; a figure may be withheld, with the reason. Waits up to 90 s, then answers status "running" with a run_id for get_result.

Input parameters:

- `account_usd` (number): The starting account, in US dollars.
- `code` (string, required): One backtesting.py Strategy class (describe_data has the rules).
- `costs`: The costs: two fractions of the trade value, or a list of costs each in its own unit.
- `label` (string): A short text shown with the run.
- `stated_rules` (array): Rules the trades are checked against, one per kind.
- `study_id` (string, required): The id create_study returned.
- `symbol` (string, required): One ticker ("QQQ", "BTCUSD", "EURUSD").
- `timeframe` (string, required)
- `window`: Dates ({from, to}; `to` may be "latest") or the last N days, months or years ({last: {n, unit}}). Left out: the standard history, the newest bars counted back from the newest bar, cut at the start of…

Output parameters:

- `basis` (object)
- `buy_and_hold` (object)
- `cost_table` (object): Every cost the run charged: each component (commission, slippage, spread) with its amount, unit and basis, what it comes to per round trip, when it is taken, and whether it is yours (caller) or the m…
- `disclaimer` (string): The hypothetical-performance line, on a result (a run, or a study's runs): the same text the app shows with every result.
- `grade` (object|null)
- `grade_withheld` (string|null)
- `headline` (string)
- `held_out` (object|null)
- `hold_out_estimate` (object)
- `lines` (array)
- `look_ahead` (string|null)
- `market` (object)
- `message` (string)
- `metrics` (object)
- `next` (object|null)
- `notices` (array)
- `poll_after_seconds` (integer)
- `reason` (string)
- `replayed_from` (string)
- `retryable` (boolean)
- `run_id` (string)
- `runs_left_today` (object)
- `settings` (object)
- `stated_rule_breaks` (array)
- `status` (string)
- `study_id` (string)
- `trade_list` (object): The run's trades in the order they were made, a page at a time: when each entered and left (a day on daily bars; a time on the `clock` on shorter ones; "open at the end" for a trade still open when t…
- `version` (integer)
- `warnings` (array)
- `window` (object)

### `get_result` (~192 tokens)

Read a result

Reads a backtest result by its run_id (when run_backtest answered "running", call this until the run ends), or a study's runs by study_id (one of them with version). A run comes with its trades, a page at a time (trades_page).

Input parameters:

- `before_version`: With study_id: the 20 runs before this version (a list's `next` gives it).
- `run_id` (string): A run_id run_backtest or run_held_out_test returned.
- `study_id` (string): A study's id, for its runs (the newest 20 by default); with `version`, one of them.
- `trades_page` (integer): With a run (run_id, or study_id with version): the page of its trade list, from 1; a page holds 100 trades. Left out: the first.
- `version`

Output parameters:

- `basis` (object)
- `buy_and_hold` (object)
- `cost_table` (object): Every cost the run charged: each component (commission, slippage, spread) with its amount, unit and basis, what it comes to per round trip, when it is taken, and whether it is yours (caller) or the m…
- `disclaimer` (string): The hypothetical-performance line, on a result (a run, or a study's runs): the same text the app shows with every result.
- `grade` (object|null)
- `grade_withheld` (string|null)
- `headline` (string)
- `held_out` (object|null)
- `highest_simulated_value` (object|null)
- `hold_out` (object|null)
- `hold_out_estimate` (object)
- `lines` (array)
- `look_ahead` (string|null)
- `market` (object)
- `message` (string)
- `metrics` (object)
- `next` (object|null)
- `notices` (array)
- `poll_after_seconds` (integer)
- `reason` (string)
- `replayed_from` (string)
- `retryable` (boolean)
- `run_id` (string)
- `runs` (array)
- `runs_left_today` (object)
- `settings` (object)
- `stated_rule_breaks` (array)
- `status` (string)
- `study_id` (string)
- `trade_list` (object): The run's trades in the order they were made, a page at a time: when each entered and left (a day on daily bars; a time on the `clock` on shorter ones; "open at the end" for a trade still open when t…
- `version` (integer)
- `warnings` (array)
- `window` (object)

### `run_held_out_test` (~128 tokens)

Test on the hold-out

The final out-of-sample check: runs a finished run's exact code once over its study's hold-out period. Call it at the end, on the run you would keep: the first held-out test per account and market is the one that counts, and a hold-out with too few trades for figures (a result needs 10) spends it all the same: each development run says how many trades the hold-out is expected to hold, in `hold_out_estimate`.

Input parameters:

- `run_id` (string, required): A run of a study with a hold-out, which ran.

Output parameters:

- `basis` (object)
- `buy_and_hold` (object)
- `cost_table` (object): Every cost the run charged: each component (commission, slippage, spread) with its amount, unit and basis, what it comes to per round trip, when it is taken, and whether it is yours (caller) or the m…
- `disclaimer` (string): The hypothetical-performance line, on a result (a run, or a study's runs): the same text the app shows with every result.
- `grade` (object|null)
- `grade_withheld` (string|null)
- `headline` (string)
- `held_out` (object|null)
- `hold_out_estimate` (object)
- `lines` (array)
- `look_ahead` (string|null)
- `market` (object)
- `message` (string)
- `metrics` (object)
- `next` (object|null)
- `notices` (array)
- `poll_after_seconds` (integer)
- `reason` (string)
- `replayed_from` (string)
- `retryable` (boolean)
- `run_id` (string)
- `runs_left_today` (object)
- `settings` (object)
- `stated_rule_breaks` (array)
- `status` (string)
- `study_id` (string)
- `trade_list` (object): The run's trades in the order they were made, a page at a time: when each entered and left (a day on daily bars; a time on the `clock` on shorter ones; "open at the end" for a trade still open when t…
- `version` (integer)
- `warnings` (array)
- `window` (object)

### `compare_trades` (~274 tokens)

Check my own trades

Checks a trader's own trades against a run's rule, on the price history this server holds: for each trade, in their order, whether the rule would have taken it and, if not, why (already in a trade, no entry signal that day, no bar that day). Use it when the trader lists trades they made and asks which the rule took, instead of working it out by hand. Takes a run (run_id) and the trades, each {side, entry, exit?}; answers in words, never prices. It re-runs the run's code, so it counts as a run on the run's symbol.

Input parameters:

- `run_id` (string): A run_id run_backtest or run_held_out_test returned: the run whose code is checked against the trades.
- `study_id` (string): A study's id; with `version`, one of its runs.
- `time_zone` (string): The clock the trader says their times are on ("UTC+2", "New York", "Europe/Prague"). Left out: the clock the run's trade list is on (ET for stocks and ETFs, UTC otherwise).
- `trades` (array, required): The trader's own trades, in their order, each stamp exactly as the trader wrote it.
- `version`

Output parameters:

- `clock` (string): On bars shorter than a day: the clock every time of the answer is on (your time_zone as you stated it, else ET for stocks and ETFs, UTC otherwise).
- `counts` (object)
- `market` (object)
- `message` (string)
- `note` (string)
- `reason` (string)
- `run_id` (string)
- `runs_left_today` (object)
- `status` (string)
- `study_id` (string)
- `summary` (string)
- `trades` (array)
- `version` (integer)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/com-quantwright-agent-api/api-mcp#diagnostics

## Score history

- 2026-11-05: 80
- 2026-11-04: 80
- 2026-11-03: 79
- 2026-11-02: 79
- 2026-11-01: 78
- 2026-10-31: 78
- 2026-10-30: 77
- 2026-10-29: 77

## Common questions

### What is the QuantWright MCP server?

QuantWright is an MCP server listed in the public MCP registry as com.quantwright/agent-api. Backtests with costs, a held-out test that counts once, a score that counts every version tried. This page covers its hosted endpoint (https://quantwright.com/api/mcp).

### Is the QuantWright MCP server safe to use?

QuantWright scores 80 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the QuantWright MCP server expose?

QuantWright exposes 6 tools: describe_data, create_study, run_backtest, get_result, run_held_out_test, compare_trades. Their descriptions and schemas cost roughly 1,237 tokens of context every time the server is loaded.

### Does the QuantWright MCP server require authentication?

Yes. QuantWright asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.

### Is the QuantWright MCP server still maintained?

QuantWright is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the QuantWright MCP server?

VerifyMCP ran 13 checks on QuantWright, covering TLS certificate, Authorization, HTTPS, HSTS header and 9 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://quantwright.com/api/mcp
- Authorisation metadata: https://quantwright.com/.well-known/oauth-protected-resource/api/mcp
- Website: https://quantwright.com/docs/agent-api
- Changelog RSS feed: https://verifymcp.io/servers/com-quantwright-agent-api/api-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-quantwright-agent-api/api-mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/com-quantwright-agent-api/api-mcp.json
- HTML version of this page: https://verifymcp.io/servers/com-quantwright-agent-api/api-mcp