# quant-swarm (pypi · quant-swarm)

Risk-first pre-trade checks, PIT data, leakage audits, and a Shadow Tournament vs a live champion.

- Trust score: 73/100 (medium)
- Change this week: −6
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `swarm-mcp-503318750546.europe-west1.run.app`: 36/100, [markdown](https://verifymcp.io/servers/blink1217-quant-swarm/mcp-data.md), [page](https://verifymcp.io/servers/blink1217-quant-swarm/mcp-data)
- remote · `swarm-mcp-503318750546.europe-west1.run.app`: 36/100, [markdown](https://verifymcp.io/servers/blink1217-quant-swarm/mcp-gym.md), [page](https://verifymcp.io/servers/blink1217-quant-swarm/mcp-gym)
- remote · `swarm-mcp-503318750546.europe-west1.run.app`: 36/100, [markdown](https://verifymcp.io/servers/blink1217-quant-swarm/mcp-warden.md), [page](https://verifymcp.io/servers/blink1217-quant-swarm/mcp-warden)
- pypi · `quant-swarm`: 73/100 (this document), [markdown](https://verifymcp.io/servers/blink1217-quant-swarm/quant-swarm.md), [page](https://verifymcp.io/servers/blink1217-quant-swarm/quant-swarm)

## Channel facts

- Registry: `pypi`
- Package: `quant-swarm`
- Version: `0.5.0`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, 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.

- **Supply Chain Security**: 50/100
  - Malware scan not yet available for this package.
  - No known CVEs affecting this package version or its production dependencies.
  - Runs setuptools.build_meta at install time, a recognised build step with no custom scripting around it.
  - 1 of 36 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 84/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to blink1217/trading-swarm-mcp).
  - License check failed: the license (MIT AND LicenseRef-Guardrails-Source-Available) isn't a recognized OSI-approved license.
  - Actively maintained (last published 5 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 81/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 2024 tokens (~119/item across 17 items; 13 tools + 4 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 63/100
  - Stability observed for 19 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% 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 13 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 15 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 quant-swarm MCP server?

quant-swarm runs locally as a PyPI package, launched with uvx quant-swarm. 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 blink1217-quant-swarm -- uvx quant-swarm
```

### Cursor

```json
{
  "mcpServers": {
    "blink1217-quant-swarm": {
      "command": "uvx",
      "args": [
        "quant-swarm"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "blink1217-quant-swarm": {
      "command": "uvx",
      "args": [
        "quant-swarm"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add blink1217-quant-swarm -- uvx quant-swarm
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "blink1217-quant-swarm": {
      "type": "local",
      "command": [
        "uvx",
        "quant-swarm"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add blink1217-quant-swarm --command uvx --arg quant-swarm
```

### Hermes

```yaml
mcp_servers:
  blink1217-quant-swarm:
    command: "uvx"
    args: ["quant-swarm"]
```

### Netclaw

```json
{
  "McpServers": {
    "blink1217-quant-swarm": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "quant-swarm"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add blink1217-quant-swarm -t stdio -c uvx -a quant-swarm
```

### Other

```json
{
  "mcpServers": {
    "blink1217-quant-swarm": {
      "command": "uvx",
      "args": [
        "quant-swarm"
      ]
    }
  }
}
```

## 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 73, +9)

- [functional improvement] Stability: unverified → 0.60

### 2026-10-30 (score 64, −15)

- [security regression] Malware scan: pass → unverified

### 2026-10-29 (score 79, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-10-26 (score 64, −15)

- [security regression] Malware scan: pass → unverified

### 2026-10-23 (score 79, 0)

- [functional regression] Security disclosure: unverified → fail

### 2026-10-22 (score 79, 0)

- [functional regression] Security disclosure: fail → unverified

### 2026-10-17 (score 79, +32)

- [security improvement] Injection markers: unverified → pass
- [security] First check of Judged manipulation: pass
- [security] Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window).
- [functional improvement] Tool coverage: unverified → 100
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Schema quality: unverified → 100
- [functional] First check of Destructive annotations: pass
- [functional] First check of Schema quality: fail
- [functional] First check of Tool coverage: 100
- [functional] First check of Schema quality: good
- [functional] First check of Tool coverage: 100
- [functional] First check of Schema quality: fail

### 2026-10-14 (score 47, 0)

- [security] Tool safety: Tool safety not yet verified: this package could not be installed in a clean environment, so we have no tool text to scan.
- [security] Stability: Stability not yet verified: this package could not be installed in a clean environment, so we have no schema to compare.
- [functional] Tool coverage: Tool coverage not yet verified: this package could not be installed in a clean environment, so we have no tool definitions to assess.
- [functional] Capabilities: Protocol version not yet verified: this package could not be installed in a clean environment, so we never saw its MCP handshake.
- [functional] Schema quality: Schema quality not yet verified: this package could not be installed in a clean environment, so we have no schema to assess.

## MCP tools (13)

### `features.build` (~130 tokens)

Build Feature Vector

Exact FEATURE_ORDER feature vector for symbol at session as_of, with provenance per field.

Runs the no-lookahead guard (every feature must equal a fresh causal recomputation at as_of)
and the point-in-time provenance guards. Tier-B/C fields without recorded point-in-time
evidence are returned as UNSCORABLE — never neutral-filled with 0.0.

Input parameters:

- `as_of` (string, required): Point-in-time session date (YYYY-MM-DD); every feature is a causal recomputation at this date, never a later revision.
- `symbol` (string, required): Ticker symbol, e.g. AAPL.

Output parameters:

- `access` (string)
- `as_of` (string)
- `error` (string)
- `escalation` (object)
- `feature_order` (array)
- `how` (string)
- `limits` (object)
- `no_lookahead` (string)
- `provenance` (array)
- `request_access_at` (string)
- `symbol` (string)
- `tool` (string)
- `vector` (object)
- `violations` (array)

### `market.pulse` (~175 tokens)

Market Pulse

Derived-only technical snapshot: trend, volatility, oscillator, volume, and regime signals.

Never returns raw open/high/low/close/volume or provider quote values — only ratios,
percentile ranks (trailing 1y), and causal regime labels. Pass `symbols` to use the shared
cache/relay, or pass your own `bars` rows (same shape as get_bars) for symbols we don't carry.
This is the recommended tool for general market analysis; not investment advice.

Input parameters:

- `bars`: Optional caller-supplied OHLCV rows [{symbol, ts, open, high, low, close, volume}] for symbols the shared cache does not carry.
- `symbols`: Optional list of tickers to analyze via the shared cache/relay (default: none — supply symbols or your own bars).

Output parameters:

- `access` (string)
- `error` (string)
- `how` (string)
- `learn_more` (string)
- `methodology` (string)
- `not_investment_advice` (boolean)
- `request_access_at` (string)
- `signals` (object)
- `source` (string)
- `tool` (string)

### `market.sentiment` (~64 tokens)

Sentiment Pulse

Derived-only sentiment/news snapshot for one symbol: headline count, earnings flag,
and day-change bucket. Never returns raw quote values or headline text — quantized signals
only. Not investment advice.

Input parameters:

- `symbol` (string, required): Ticker symbol, e.g. TSLA.

Output parameters:

- `access` (string)
- `as_of` (string)
- `day_change_bucket` (string)
- `earnings_within_3d` (boolean)
- `error` (string)
- `headline_count_7d` (integer)
- `how` (string)
- `learn_more` (string)
- `not_investment_advice` (boolean)
- `provenance` (object)
- `request_access_at` (string)
- `symbol` (string)
- `tool` (string)

### `market.climate` (~207 tokens)

Operating-Area Climate Research

Area-specific weather research for each symbol's operating footprint.

Fetches the keyless Open-Meteo 14-day forecast AND the same 14 calendar days one
year prior per operating area (where the product is made and/or sold), then
exposure-weights into symbol aggregates. Returns derived sums/means/ratios/
counts/buckets per area and per symbol — never raw per-day series. A London
forecast is never substituted for a product made and sold in New York. Free tool;
no provider credentials needed. Not investment advice.

Input parameters:

- `areas`: Optional caller-supplied operating areas for underlyings outside the shipped registry: [{symbol, name?, lat, lon, weight?}]. Weather is fetched ONLY for these made/sold geographies — never the listin…
- `symbols` (array, required): Tickers whose operating-area climate to research (registry-covered weather-exposed underlyings, or any ticker you supply `areas` for).

Output parameters:

- `access` (string)
- `climate` (object)
- `covered` (array)
- `error` (string)
- `how` (string)
- `learn_more` (string)
- `methodology` (string)
- `not_investment_advice` (boolean)
- `registry_coverage` (integer)
- `request_access_at` (string)
- `source` (string)
- `tool` (string)
- `uncovered` (array)

### `pit.analysis` (~284 tokens)

PIT Significance Explained

Is this alternative-data signal real for these tickers - or noise?

Explained PIT significance per underlying, from the release-gated
pit_analysis/v2 run: every (dataset|metric|horizon) family reports rolling
52-week IR, Newey-West two-sided p, observation count, research-only and
staleness flags, and an explicit verdict - SIGNIFICANT (survives the funnel's
IR > 0 and Benjamini-Hochberg q <= 0.05 gate of the same run), NOT_SIGNIFICANT,
below the observation floor, or UNVERIFIED. Point-in-time, derived-only, no raw
values; free lite answers, Pro gives every family. Not investment advice.

Input parameters:

- `as_of`: Point-in-time cutoff (YYYY-MM-DD, default today): release points after this date are ignored.
- `horizons`: Subset of the run's horizons, e.g. ['1d','1w','1m','6m'] (default: all).
- `max_families` (integer): Max families returned per underlying, most significant first (1-100).
- `only_significant` (boolean): Only return families with at least one SIGNIFICANT horizon (counts still cover everything).
- `symbols` (array, required): Underlying tickers to query (max 10). Results are per underlying.

Output parameters:

- `access` (string)
- `error` (string)
- `how` (string)
- `how_to_read` (string)
- `learn_more` (string)
- `methodology` (string)
- `not_investment_advice` (boolean)
- `request_access_at` (string)
- `run_id` (string)
- `source` (string)
- `tool` (string)
- `unavailable` (object)
- `underlyings` (object)

### `cache.warm` (~110 tokens)

Warm Cache

Backfill history for a universe (default: the swarm's 13-symbol universe).

After warming, every finalized session is immutable in cache — replays cost zero API calls.
Requires network and a valid SWARM_MCP_ACCESS_TOKEN; refused in cache.offline mode.

Input parameters:

- `universe`: Optional list of tickers to backfill (default: the swarm's 13-symbol universe).
- `years` (number): Years of daily history to backfill (max limited by the plan's backfill window).

Output parameters:

- `access` (string)
- `error` (string)
- `escalation` (object)
- `how` (string)
- `immutability` (string)
- `limits` (object)
- `request_access_at` (string)
- `rows_upserted` (integer)
- `symbols` (object)
- `tool` (string)

### `cache.stats` (~26 tokens)

Cache Stats

Cache inventory: rows per symbol, session bounds, depth in weeks, API-call accounting.

Output parameters:

- `access` (string)
- `error` (string)
- `how` (string)
- `offline_mode` (boolean)
- `per_symbol` (object)
- `request_access_at` (string)
- `tool` (string)

### `cache.offline` (~50 tokens)

Toggle Offline Mode

Toggle cache-only mode: no network access; reads served from cache, fetches refused.

Input parameters:

- `enabled` (boolean, required): True = cache-only mode (no network; reads from cache, fetches refused).

Output parameters:

- `access` (string)
- `enabled` (boolean)
- `error` (string)
- `how` (string)
- `request_access_at` (string)
- `semantics` (string)
- `tool` (string)

### `market.regime` (~100 tokens)

Regime Snapshot

Free tool — causal regime label per symbol (high_vol / drawdown / chop /
melt_up / gap_heavy), derived from point-in-time bars with no lookahead.

Input parameters:

- `bars`: Optional caller-supplied OHLCV rows [{symbol, ts, open, high, low, close, volume}] for symbols the shared cache does not carry.
- `symbols`: Optional list of tickers to label via the shared cache/relay.

Output parameters:

- `access` (string)
- `error` (string)
- `how` (string)
- `learn_more` (string)
- `methodology` (string)
- `not_investment_advice` (boolean)
- `regimes` (object)
- `request_access_at` (string)
- `source` (string)
- `tool` (string)

### `market.microstructure` (~93 tokens)

Microstructure Snapshot

Derived-only microstructure snapshot: order-flow imbalance, VPIN toxicity
bucket, and effective half-spread — percentile-ranked within the series.
Pro tool.

Input parameters:

- `bars`: Optional caller-supplied OHLCV rows [{symbol, ts, open, high, low, close, volume}].
- `symbol`: Ticker to analyze (served from the shared cache/relay); omit when passing your own bars.

Output parameters:

- `access` (string)
- `error` (string)
- `how` (string)
- `learn_more` (string)
- `methodology` (string)
- `not_investment_advice` (boolean)
- `request_access_at` (string)
- `signals` (object)
- `source` (string)
- `tool` (string)

### `volume.forecast` (~84 tokens)

Volume Forecast

Derived-only volume forecast expressed as a ratio to the trailing 20-session
average, with a 10th-90th percentile interval. Never returns raw share counts.
Pro tool.

Input parameters:

- `horizon_bars` (integer): Number of sessions ahead to forecast (1-10).
- `symbol` (string, required): Ticker to forecast (served from the shared cache/relay).

Output parameters:

- `access` (string)
- `as_of` (string)
- `error` (string)
- `forecast_ratio_to_20d_avg` (number)
- `horizon_bars` (integer)
- `how` (string)
- `interval_p10_p90` (array)
- `learn_more` (string)
- `not_investment_advice` (boolean)
- `note` (string)
- `request_access_at` (string)
- `symbol` (string)
- `tool` (string)

### `market.screen` (~105 tokens)

Screen Universe

Ranked, derived-only screen over the symbol list — cross-sectional
percentile ranks, never raw prices or volumes. Pro tool.

Input parameters:

- `bars`: Optional caller-supplied OHLCV rows [{symbol, ts, open, high, low, close, volume}].
- `criteria` (string): Screen criterion: breakout | momentum | volume | anomaly.
- `symbols`: List of tickers to screen (via the shared cache/relay); omit when passing your own bars.

Output parameters:

- `access` (string)
- `criteria` (string)
- `error` (string)
- `how` (string)
- `learn_more` (string)
- `methodology` (string)
- `not_investment_advice` (boolean)
- `request_access_at` (string)
- `screened` (object)
- `source` (string)
- `tool` (string)

### `market.rank` (~115 tokens)

Cross-Sectional Rank

Cross-sectional percentile rank of one derived metric across the symbol
list — no raw metric values returned. Pro tool.

Input parameters:

- `bars`: Optional caller-supplied OHLCV rows [{symbol, ts, open, high, low, close, volume}].
- `metric` (string): Metric to rank by: momentum_20d | volume_ratio_20d | atr_pct | rsi_14.
- `symbols`: List of tickers to rank (via the shared cache/relay); omit when passing your own bars.

Output parameters:

- `access` (string)
- `error` (string)
- `how` (string)
- `learn_more` (string)
- `methodology` (string)
- `metric` (string)
- `not_investment_advice` (boolean)
- `rank` (object)
- `request_access_at` (string)
- `source` (string)
- `tool` (string)

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/blink1217-quant-swarm/quant-swarm#diagnostics

## Score history

- 2026-11-05: 73
- 2026-11-04: 73
- 2026-11-03: 64
- 2026-11-02: 64
- 2026-11-01: 64
- 2026-10-31: 64
- 2026-10-30: 64
- 2026-10-29: 79
- 2026-10-28: 64
- 2026-10-27: 64
- 2026-10-26: 64
- 2026-10-25: 79
- 2026-10-24: 79
- 2026-10-23: 79
- 2026-10-22: 79
- 2026-10-21: 79
- 2026-10-20: 79
- 2026-10-19: 79
- 2026-10-18: 79
- 2026-10-17: 79
- 2026-10-16: 47
- 2026-10-15: 47
- 2026-10-14: 47
- 2026-10-13: 47
- 2026-10-12: 32
- 2026-10-08: 32

## Common questions

### What is the quant-swarm MCP server?

quant-swarm is an MCP server listed in the public MCP registry as io.github.blink1217/quant-swarm. Risk-first pre-trade checks, PIT data, leakage audits, and a Shadow Tournament vs a live champion. This page covers its PyPI package (quant-swarm).

### Is the quant-swarm MCP server safe to use?

quant-swarm scores 73 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 5 November 2026. Its build provenance is signed and verified. 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 quant-swarm MCP server expose?

quant-swarm exposes 13 tools: features.build, market.pulse, market.sentiment, market.climate, pit.analysis, and 8 more. Their descriptions and schemas cost roughly 1,543 tokens of context every time the server is loaded.

### Is the quant-swarm MCP server still maintained?

quant-swarm 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.

## Links

- PyPI project: https://pypi.org/project/quant-swarm/
- Socket report: https://socket.dev/pypi/package/quant-swarm
- Repository: https://github.com/blink1217/trading-swarm-mcp
- Website: https://1.21initiative.com/mcp/
- Changelog RSS feed: https://verifymcp.io/servers/blink1217-quant-swarm/quant-swarm.xml
- Changelog JSON feed: https://verifymcp.io/servers/blink1217-quant-swarm/quant-swarm.json
- HTML version of this page: https://verifymcp.io/servers/blink1217-quant-swarm/quant-swarm