Boltwork Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Boltwork Mcp (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
mcp-name: io.github.Squidboy30/boltwork-mcp
MCP server for Boltwork — AI services that pay for themselves via Bitcoin Lightning.
Give your AI agent PDF summarisation, code review, translation, web extraction, document comparison, and persistent memory — all paid autonomously in sats. No API keys. No subscriptions. No accounts.
Boltwork is a pay-per-call AI services API that uses the L402 protocol — your agent makes a request, receives a Lightning invoice, pays it automatically, and gets the result back. No human involved.
This package wraps Boltwork as an MCP server so any MCP-compatible AI (Claude, Cursor, Windsurf, etc.) can use it as a tool — with payments handled transparently in the background.
Two tools work right now with zero setup:
{
"mcpServers": {
"boltwork": {
"command": "uvx",
"args": ["boltwork-mcp"],
"env": {}
}
}
}Then ask your AI:
"Use trial_review_code to review this: def add(a, b): return a + b"
"Use trial_summarise to summarise this: <paste any text>"Real AI results instantly. No Lightning wallet. No setup. Rate limited to 5 calls/hour.
pip install boltwork-mcp
# If using NWC (Alby, Mutiny, Coinos, etc.):
pip install "boltwork-mcp[nwc]"Or use directly with uvx — no install needed:
uvx boltwork-mcpFour wallet backends are supported. Pick whichever fits your setup:
Works with Alby, Mutiny Wallet, Coinos, Primal, Cashu.me, and any NWC-compatible wallet.
{
"mcpServers": {
"boltwork": {
"command": "uvx",
"args": ["boltwork-mcp"],
"env": {
"NWC_CONNECTION_STRING": "nostr+walletconnect://your-string-here"
}
}
}
}Requires: pip install "boltwork-mcp[nwc]"
Works with lnbits.com or any self-hosted LNbits instance. Popular with BTCPay Server users and home node operators.
{
"mcpServers": {
"boltwork": {
"command": "uvx",
"args": ["boltwork-mcp"],
"env": {
"LNBITS_URL": "https://lnbits.com",
"LNBITS_API_KEY": "your-invoice-key-here"
}
}
}
}For self-hosted: set LNBITS_URL to your instance URL (e.g. https://lnbits.yourdomain.com).
Works with a Strike account. Custodial, simple API key setup. Good for US users or anyone who already uses Strike.
{
"mcpServers": {
"boltwork": {
"command": "uvx",
"args": ["boltwork-mcp"],
"env": {
"STRIKE_API_KEY": "your-api-key-here"
}
}
}
}Works with Phoenixd — ACINQ's self-hosted Lightning node. Simple REST API, no channel management.
{
"mcpServers": {
"boltwork": {
"command": "uvx",
"args": ["boltwork-mcp"],
"env": {
"PHOENIXD_URL": "http://localhost:9740",
"PHOENIXD_PASSWORD": "your-password-here"
}
}
}
}| Client | Config file |
|---|---|
| Claude Desktop (macOS) | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Claude Desktop (Windows) | %APPDATA%\Claude\claude_desktop_config.json |
| Cursor | .cursor/mcp.json in your project |
| Windsurf | ~/.codeium/windsurf/mcp_config.json |
| Tool | What it does | Cost |
|---|---|---|
trial_summarise | Summarise text — free trial | Free |
trial_review_code | Review code — free trial | Free |
summarise_pdf | Summarise a PDF from URL | 500 sats |
summarise_webpage | Summarise any web page | 100 sats |
review_code | Full code review with bugs, security, quality | 2000 sats |
review_code_url | Review code from GitHub/GitLab URL | 2000 sats |
extract_data | Extract structured data from PDF | 200 sats |
translate | Translate text or document (24 languages) | 150 sats |
extract_tables | Extract all tables from a PDF | 300 sats |
compare_documents | Diff two PDFs | 500 sats |
explain_code | Explain code in plain English | 500 sats |
memory_store | Store persistent agent memory | 10 sats |
memory_retrieve | Retrieve agent memory | 5 sats |
memory_delete | Delete a memory key | Free |
run_workflow | Chain multiple services in one call | 1000 sats |
When your agent calls a paid tool:
Your agent never sees this — it just gets the result.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.