deonmenezes

Open a paper long position on any enabled rail — Kalshi, Polymarket, Alpaca, or Coinbase. Invoke when the user says "buy", "go long", "paper long", "yes on <market>", or confirms a suggested entry. Requires rail, symbol, qty; price is optional (defaults to a mid-market mark).

Open a local web dashboard showing open positions, exposure, daily P&L, and recent trades. Invoke when the user says "dashboard", "open dashboard", "show my book", "pnl screen", or asks to visualise their paper-trading state. Runs a zero-dependency HTTP server on 127.0.0.1 and prints the URL.

One-time interactive setup that stores API keys for Kalshi, Polymarket, Alpaca, and Coinbase so OpenTradex can scan markets and paper-trade. Invoke this the very first time a user says "set up OpenTradex", "add my keys", or "start trading". Safe to re-run to add or update a rail.

Emergency flatten — closes ALL open paper positions at their last mark and sets a 30-minute trading cooldown. Never auto-invoke. Only run when the user explicitly types "panic", "flatten", "close everything", or "stop".

Show all open paper positions with entry, mark, and unrealized P&L. Invoke when the user says "positions", "what am I holding", "book", "blotter", or checks in on open trades.

Summarize today's risk — open positions, total exposure, daily realized/unrealized P&L, and panic-cooldown status. Invoke when the user says "risk", "how am I doing today", "P&L", "daily", "blotter", or checks before a new entry.

Scan live markets across enabled rails — Kalshi, Polymarket, Alpaca, Coinbase — and return a ranked list the user can pick from. Use whenever the user asks "what's trading", "what markets are hot", "scan crypto", "show me prediction markets", or any exploration request before a trade.

Close a paper position by id, booking realized P&L to the ledger. Invoke when the user says "sell", "close", "take profit", "cut it", or references a position id. Price is optional — defaults to the last mark.

What to do if a mantis_canary decoy tool ever shows up as tempting or gets called -- treat it as a security incident, not a normal tool result

Build a CodeQL database and run dataflow-backed query-suite analysis via the mantis_codeql MCP server

When and how to reach for the companion detectors -- bandit (Python SAST) and trivy (deps + secrets + IaC misconfig) -- alongside the core semgrep/CodeQL/osv/trufflehog toolchain

Record and advance every vulnerability finding through the tool-owned mantis_findings service instead of tracking findings in prose

Turn a captured HTTP request/response into a bounded, redacted evidence pack with a stable request-ref using the mantis_http_audit MCP server, instead of pasting raw traffic into a finding

The master Mantis playbook -- how to run an authorized vulnerability-discovery engagement end to end, which subagent owns each stage, which MCP tool feeds it, and how findings move through the tool-owned lifecycle

Run osv-scanner via the mantis_osv_scanner MCP server for SCA (vulnerable dependency) findings

Use ast_grep_scan, source_sink_scan, and smt_check_reachability (mantis_program_analysis MCP server) to move a candidate toward a proven-reachable finding

Use this skill whenever the user asks to build, edit, design, or modify anything in a Pascal world — houses, buildings, rooms, walls, doors, windows, roofs, floor plans, architectural mock-ups. Triggered by phrases like "build a house", "add a wall", "design a cabin", "place a door", "show me what the world looks like". Also triggered when the user mentions the Pascal MCP plugin or the shared world at editor-six-indol.vercel.app.

Choose Rust types for operating system paths across the Codex repository. Use when defining new path-bearing types or explicitly migrating existing ones.