BotShelf Vampire BOTSHELF VAMPIRE Register

EXPLORE · SAME JOB, BY TOOL

Deep Research: 7 versions, one per tool

The same job, written separately for CrewAI, LangGraph, LM Studio, MCP, n8n, Ollama, Open WebUI. Each version is its own text; what you get is text to copy into that tool, not something BSV installs or runs for you. Pick the tool you already use and open its page: the full text is there, with Copy and a file to download. No sign-up.

ToolDescription on the item pageStatusRaw
CrewAITurn a fuzzy question into a sourced research brief with open questions and next checks — not a final verdict. (CrewAI multi-agent Python sketch for one job.)Untestedfile
LangGraphTurn a fuzzy question into a sourced research brief with open questions and next checks — not a final verdict. (LangGraph stateful graph with a human checkpoint.)Untestedfile
LM StudioTurn a fuzzy question into a sourced research brief with open questions and next checks — not a final verdict. (LM Studio system prompt + local server notes.)Untestedfile
MCPTurn a fuzzy question into a sourced research brief with open questions and next checks — not a final verdict. (MCP-oriented tool + prompt recipe (read-only first).)Untestedfile
n8nTurn a fuzzy question into a sourced research brief with open questions and next checks — not a final verdict. (n8n workflow recipe (Manual Trigger → LLM → output).)Untestedfile
OllamaTurn a fuzzy question into a sourced research brief with open questions and next checks — not a final verdict. (Ollama Modelfile + chat prompt for offline runs.)Check recordedfile
Open WebUITurn a fuzzy question into a sourced research brief with open questions and next checks — not a final verdict. (Open WebUI system prompt / custom agent notes.)Untestedfile

1 of these 7 versions have a recorded check; the rest are marked untested. Status comes from each item page. BSV does not run any of them for you.

Search this job in Explore How to run How to fork All jobs

Used one of these versions? Say how it went, or ask, in Community.