# Big Boss MCP > Use when a user asks for live BIG BOSS BOT intelligence, situational awareness, recent conflict updates, specific BIG BOSS BOT feed data, or searches across the Ukraine or Middle East theaters through MCP. - Skill: `tomevault-io/big-boss-mcp` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/big-boss-mcp` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/big-boss-mcp/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/big-boss-mcp --- # BIG BOSS BOT MCP Use this skill when the user wants live intel from BIG BOSS BOT rather than static knowledge. ## Tool choice - Use `get_snapshot` for broad questions like “What’s happening in Ukraine?” or “Give me a theater overview.” - Use `search_intel` for mention searches, recent sightings, and “find reports about X” requests. - Use `get_feed` when the user wants one exact dataset such as `flights`, `news`, `alerts`, or `polymarket`. ## Theater selection - Use `ukraine` for Ukraine, Russia, Crimea, Black Sea, Kyiv, Kharkiv, Odesa, Kursk, or Belgorod requests. - Use `middle-east` for Israel, Iran, Gaza, Lebanon, Syria, Red Sea, Gulf, Yemen, or Hormuz requests. - If the user is ambiguous, ask for the theater or clearly state the assumption you made. ## Response rules - Treat MCP output as the source of truth. - Cite the MCP tool and feed names you used. - Preserve timestamps and source labels when they are available. - Do not invent feed items or summarize beyond what the MCP output supports. ## Setup If the MCP server is not available yet, read [references/setup.md](references/setup.md). --- > Source: [kitakitsune0x/bigbossbot](https://github.com/kitakitsune0x/bigbossbot) — distributed by [TomeVault](https://tomevault.io).