mcp-utility-server
Provides trading signals by fetching OHLCV data from Binance and computing RSI and moving average crossovers, outputting BUY/SELL/HOLD recommendations.
Allows scanning Docker container logs to detect trades, errors, and signals, using the Docker CLI.
Provides tools for sending Telegram polls and messages, and for notifying users of page changes via Telegram.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-utility-server What's the trading signal for BTCUSDT?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Utility Server
A production-grade Model Context Protocol server with 10 utility tools for daily operations. Built with FastMCP, stdlib-first, fully tested (TDD).
Why this server: it demonstrates real MCP engineering — not a hello-world. It includes:
Retry/backoff with jitter on all external HTTP calls (429/5xx handling)
State persistence for page monitoring (SHA-256 diff tracking)
Observability — per-tool retry counters exposed as an MCP tool
Sandboxing for file operations (optional
FILE_SANDBOX)Deterministic outputs (word-of-day selection is hash-based, same all day)
~30 pytest tests with mocked network (no external deps in CI)
Tools
Tool | Description |
| Convert MD/HTML/CSV/JSON/Text formats (stdlib-only) |
| Poll a URL, SHA-256 diff, optional Telegram notify on change |
| Send Telegram polls (2-10 options) |
| Binance OHLCV → RSI(14) + SMA20/SMA50 cross → BUY/SELL/HOLD |
| Parse Freqtrade backtest JSON → markdown digest |
| German word (A1/A2) with translations, IPA, examples — deterministic per date |
| Daily language card: German word + SOC term (EN/DE/RU) |
| Open-Meteo threshold alerts (wind/rain) for configured cities |
| Scan trading logs for trades/errors/signals + resonance score |
| Retry/backoff observability snapshot |
Related MCP server: agent-utils-mcp
Quick start
# Install
uv sync
# Run (stdio transport — default for MCP clients)
uv run server.py
# Run (SSE transport — for remote clients)
uv run server.py --transport sse
# Test
uv run pytestConfiguration (env vars)
Env | Default | Purpose |
| — | Required for |
| — | Default chat for Telegram tools |
| unset (allow-all) | Restrict |
|
| Page-monitor state location |
|
| Max retries on 429/5xx (1 initial + N retries) |
|
| Base backoff seconds (exponential, ±20% jitter) |
|
| Retry stats persistence |
|
| Logging level |
|
|
|
Example: Claude Desktop
{
"mcpServers": {
"mcp-utility": {
"command": "uv",
"args": ["run", "--directory", "/path/to/mcp-utility-server", "server.py"]
}
}
}Example: trade signal
echo '{"symbol":"BTCUSDT","interval":"1d"}' | uv run server.py
# → {"status":"ok","symbol":"BTCUSDT","rsi":42.3,"ma_fast":42100.0,
# "ma_slow":41800.5,"signal":"BUY","reasons":["MA20 > MA50"],...}Security notes
file_convertaccepts any path whenFILE_SANDBOXis unset — always set it in productionTelegram token is read from env only, never logged
antimatter_scan("docker:...")executesdocker logs— requires docker socket access
License
MIT
Available Tools
10 toolsantimatter_scanA
Scan Freqtrade logs for trades / errors / signals.
logs_path: file path (e.g. /tmp/freqtrade.log) OR "docker:container_name" Returns dict: {status, source, lines_scanned, trades_count, buy_count, sell_count, errors_count, signals_count, recent_errors, recent_signals, score, error?}
Inputs: logs_path (required), max_errors. Returns: result
backtest_reportA
Parse Freqtrade backtest-result JSON → markdown digest.
Returns dict: {status, strategy, timeframe, trades_total, profit_pct, winrate_pct, avg_profit_pct, top_trades_md, hourly_md, error?}
Inputs: file_path (required), top_n. Returns: result
file_convertA
Convert between MD/HTML/CSV/JSON/Text formats.
Supported: md -> html | txt | json html-> md csv -> json | md_table json-> csv | md_table | pretty txt -> md
Returns dict: {status, output_path, output_format, preview, error?}
Inputs: fmt (required), path (required). Returns: result
mcp_statsA
observability — Snapshot of in-memory retry counter per tool.
Returns dict: {status, stats:{tool: {retries, events, last_status, last_error, last_at}}, total_retries, total_events, stats_path, snapshot_at} Same data is also persisted to the stats JSON file for cross-process visibility.
Inputs: no named parameters. Returns: result
page_monitorA
Check URL for changes; persist hash; notify Telegram on change.
State stored in ~/.hermes/monitors/<sha256(url)[:16]>.json. Returns dict: {status, url, hash, is_first_check, changed, diff_summary, sent_to_telegram, error?}
Inputs: url (required), chat_id, selector. Returns: result
soc_lang_cardA
Daily language card: German word (A1/A2) + SOC term (EN/DE/RU).
level: A1 | A2 focus: optional context filter: 'SOC Fundamentals' | 'Detect' | 'Triage' | 'Correlate' | 'Respond' | 'Visualize' | 'Notify' | 'Interview' Selection is deterministic (hash of date) — same card all day. Returns dict: {status, selected_for_date, german_word, soc_term, fetched_at}
Inputs: focus, level. Returns: result
telegram_pollC
Send a Telegram Poll to configured/default chat.
Returns dict: {status, message_id, poll_id, chat_id, question, error?}
Inputs: options (required), question (required), chat_id, is_anonymous. Returns: result
trade_signalA
Binance OHLCV → RSI(14) + SMA20/SMA50 cross → BUY/SELL/HOLD.
symbol examples: BTCUSDT, ETHUSDT, SOLUSDT interval: 1h | 4h | 1d | 1w Returns dict: {status, symbol, interval, price, rsi, ma_fast, ma_slow, ma_signal, signal, error?}
Inputs: symbol, interval. Returns: result
weather_alertA
Check weather for cities; return alerts when thresholds exceeded.
Default cities: Sumy, Lviv, Budapest, Vienna. Open-Meteo (no API key required). Returns dict: {status, cities_checked, alerts:[{city,condition,value,threshold,severity}], errors}
Inputs: cities, rain_threshold_mmh, wind_threshold_kmh, temp_change_threshold. Returns: result
word_of_dayA
German word of the day with translations + IPA + examples.
level: A1 | A2 Selection is deterministic (hash of date+level) — same word all day. Returns dict: {status, word, level, pos, translations, examples, pronunciation_url, fetched_at}
Inputs: level. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 10 tools
Most tools target clearly distinct functions such as file conversion, Telegram polls, trading signals, weather alerts, and log scanning. The main ambiguity is between word_of_day and soc_lang_card, since both serve daily German A1/A2 content and soc_lang_card largely supersedes word_of_day.
All names use lowercase snake_case, but the pattern is inconsistent: some are noun_action (file_convert, page_monitor), while others are noun_noun or phrase-like (mcp_stats, soc_lang_card, word_of_day). The names are readable but not highly predictable.
Ten tools is a reasonable count for a general utility server and none are clearly redundant. However, the server has no unified scope—monitoring, trading, language learning, weather, and observability are all mixed together—so the count fits numerically but the scope is broad.
Individual utilities are mostly self-contained, but there are notable gaps: page_monitor lacks monitor management, weather_alert has no city configuration, and the trading/log tools are one-shot analyzers without related operations. Because the server's domain is generic 'utilities,' completeness is hard to establish.
Maintenance
Related MCP Connectors
26 deterministic tools: PDF, image, QR, CSV/Excel, HTML to PDF, DNS/RDAP, FX rates, time zones.
26 deterministic tools: PDF, image, QR, CSV/Excel, HTML to PDF, DNS/RDAP, FX rates, time zones.
185 tools: company jobs, token/wallet/company/website reports, crypto, EVM, SEC, FRED, weather, DNS.
Bank statement PDF to CSV, OCR, Office to PDF, PDF to Word, any file to Markdown, merge and split.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceA comprehensive collection of 103 tools providing capabilities for text processing, data analysis, web development, and business management in a single server. It is designed for rapid deployment to Vercel and integrates seamlessly with MCP clients like Claude and Cursor to automate diverse workflows.-
- FlicenseAqualityCmaintenanceSwiss-army-knife utility MCP server for AI agents. 18 tools for JSON validation/formatting, base64 encode/decode, hash generation, UUID generation, URL parsing, regex testing, markdown↔HTML conversion, text stats, slug generation, datetime conversion, cron parsing, text diffing, CSV↔JSON conversion, and JWT decoding. Zero API Key required185-
- AlicenseBqualityNot gradedmaintenanceAll-in-one MCP toolkit with 29 tools across 5 modules: crypto prices, web utilities, social media, finance tracking, and productivity. MIT licensed.291-
- AlicenseNot gradedqualityCmaintenanceExposes 24 deterministic developer utilities as MCP tools — encoders and hashers, UUID/ULID and timestamp helpers, color, CSS, cron and permission converters, JSON/YAML/XML/SQL/TypeScript transformers, and a web page-to-markdown fetcher. Runs locally over stdio so agents get byte-exact results instead of approximating computations, with only the markdown fetcher making network requests.MIT