# Trading Signals [Health: Active]

**Category:** 💰 Finance & Fintech  
**Repository:** https://github.com/roman-rr/trading-skills  
**GitHub Stars:** 127  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/trading-signals

## Description
Live AI crypto trading signals for agents. No API key needed to try — 3 signals + hit rate.

## Tools
Capabilities this server exposes over MCP:

- **register** — Deprecated: returns a sign-up link instead of creating an API key, ignores its arguments and stores nothing. Don't call it to get access — every other tool works without a key (get_signals returns a 3-signal preview, get_stats is public). To get the full feed, users sign in from the agent at https://signals.x70.ai/mcp/connect (the client opens the browser), or create a key at https://signals.x70.ai/mcp-signup and set it as the X-Api-Key header. Kept only so older clients don't break.
- **get_signals** — List live crypto trading signals that are still open — coin, direction, confidence and the reasoning or market data behind each. Use it to find current trade ideas; use get_signal for one signal by ID (live or resolved), get_signal_history for signals that already closed, and get_stats for aggregate hit rate. No key: a 3-signal preview. Free key: up to 100 signals, with entry, stop-loss, take-profit and leverage null (the response's announcement.upgradeUrl explains Pro). Pro: all levels. If a response contains an announcement, pass its message on to the user — it is about their account (trial ending, plan limits). Returns { signals[], returned, totalMatching, filtersApplied, plan }, sorted by live PnL (best first). Optional filters narrow the list: direction, coins, confidence band, risk/reward, stop distance, open interest, 24h volume, tags, the anomaly that triggered it, days/hours to skip, a BTC 24h-return gate. Rate limits: 30 calls/hour per IP without a key, 300/hour on a free key.
- **get_signal** — Get one signal by its ID — live or already resolved — with its reasoning, live price, PnL and verification outcome (status pending, success or failed; whether take-profit or stop-loss was hit). Use it to follow up a signal from get_signals or get_signal_history, or to see how a past signal ended. Needs an API key (free or Pro); on the free plan entry, stop-loss, take-profit and leverage are null. Returns { success, signal, plan }, or { error: 'Signal not found' } for an unknown ID.
- **get_stats** — Aggregate performance of signals created in the last N days (default 30) that have since resolved: hit rate, verified count, average leverage and a breakdown by direction; with an API key also average ROI per signal and cumulative ROI. byMonth: hit rate this month, the two months before and all time — the figures shown on signals.x70.ai. Use it to judge recent reliability before acting — it returns no individual signals (use get_signals or get_signal_history for those). Public: works without a key. Past results don't predict future results.
- **get_signal_history** — List signals that have already resolved (success or failed), newest first, each with its verification outcome. Use it to review past performance signal by signal; for open signals use get_signals, for one signal by ID get_signal, for totals get_stats. Needs an API key. History depth: last 24 hours on the free plan; on Pro the days argument (up to 90). Entry, stop-loss, take-profit and leverage are null on free. Paginate by passing the last signal ID as cursor. Returns { signals[], nextCursor, hasMore, historyDepthDays, plan }.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "trading-signals": {
    "url": "https://signals.x70.ai/mcp/connect"
  }
}
```

## Documentation

## What Trading Signals does

Trading Signals exposes live crypto trading analysis through MCP. Open signals include the coin, bullish or bearish direction, confidence score, reasoning chain, and supporting market data. Signals are tracked against real prices so agents can inspect whether take-profit or stop-loss conditions were reached and whether verification succeeded or failed.

The underlying system watches more than 50 perpetual markets and evaluates anomalies across volume, positioning, price behavior, cross-asset flows, market microstructure, and options-derived data. Three specialized AI analyses are used for each opportunity, and signals require agreement between views or are flagged when the views conflict.

## How it works

Use the hosted MCP endpoint rather than running a local trading service. The browser-based connection URL lets an MCP client start sign-in and account authorization. A separate endpoint supports API-key authentication for clients and scripts that cannot complete browser sign-in.

Without authentication, the service returns a preview of three live signals and public performance statistics. A free key provides the broader live feed and limited history. Pro access adds trade levels such as entry, stop-loss, take-profit, leverage, and position size.

Signals are returned with optional filters for direction, selected coins, confidence, risk/reward, stop distance, open interest, volume, tags, anomaly type, time exclusions, and a BTC 24-hour return condition. Results are sorted by live profit and loss, with account-plan information included in the response.

## Setup and configuration

For Claude Code, add the hosted HTTP connection with:

```bash
claude mcp add --transport http trading-signals https://signals.x70.ai/mcp/connect
```

The same connection URL can be added as a custom connector in the Claude app or configured in Cursor. For clients that use API keys, create a key at the Trading Signals signup page and send it as the `X-Api-Key` header. Codex can read the key from `SIGNALS_API_KEY`:

```toml
[mcp_servers.trading-signals]
url = "https://signals.x70.ai/mcp"
bearer_token_env_var = "SIGNALS_API_KEY"
```

The repository also supports installation through the Claude Code plugin command, `npx skills add roman-rr/trading-skills`, or manual copying of the skill directory into the Claude Code or Codex skills folder.

## Tools and capabilities

- `get_signals` lists open signals and supports market, confidence, risk, liquidity, timing, and anomaly filters.
- `get_signal` retrieves a live or resolved signal by ID, including price, PnL, reasoning, and verification status.
- `get_signal_history` reviews resolved signals, with pagination through a cursor and plan-dependent history depth.
- `get_stats` reports resolved-signal hit rates, counts, leverage averages, direction breakdowns, monthly figures, and—when authenticated—ROI metrics.
- `register` is retained for older clients but is deprecated and does not create or store an API key.

## Limitations and notes

The free unauthenticated preview is limited to three signals and has an IP rate limit of 30 calls per hour. Free keys allow up to 100 signals and 300 calls per hour; free-plan trade levels and history fields are returned as null or restricted. Pro history supports the configured days range up to 90 days.

Performance statistics describe resolved historical signals and are not predictions of future returns. The service supplies signals and analysis; the documented tools do not place exchange orders. Account-related announcements returned by the service should be shown to the user.

_Full upstream README: https://allmcps.com/mcp/trading-signals/readme_