Stats
Actions
Tags
From ml-intern
Force a literature-first research crawl — delegates immediately to the `research` subagent without doing anything else.
How this command is triggered — by the user, by Claude, or both
Slash command
/ml-intern:research <topic, paper, or task to research>The summary Claude sees in its command listing — used to decide when to auto-load this command
Delegate this research task to the `research` subagent **immediately**. Do not attempt the research yourself — the subagent has its own context window and returns a structured recipe table. Use the Task tool with `subagent_type: "research"`. Brief: > Literature crawl for: $ARGUMENTS > > Start from anchor paper(s). Crawl citation graph for recent downstream > papers. Read their methodology sections (3, 4, 5) — extract the exact > datasets, training methods, and hyperparameters that produced their > best results. Attribute every finding to a specific result. Also find > working code example...
Delegate this research task to the research subagent immediately. Do not
attempt the research yourself — the subagent has its own context window and
returns a structured recipe table.
Use the Task tool with subagent_type: "research". Brief:
Literature crawl for: $ARGUMENTS
Start from anchor paper(s). Crawl citation graph for recent downstream papers. Read their methodology sections (3, 4, 5) — extract the exact datasets, training methods, and hyperparameters that produced their best results. Attribute every finding to a specific result. Also find working code examples using current TRL/Transformers APIs. Validate any datasets via
hf_inspect_dataset.
When the subagent returns, summarize the top recipe to the user with direct HF Hub URLs and the arxiv ID of the source paper.
npx claudepluginhub guynachshon/claude-code-ml-intern --plugin ml-intern/researchPerforms adaptive deep web research on a query with configurable --depth and --strategy options. Outputs markdown report with executive summary, analysis, confidence scores, and cited sources.
/researchPerforms structured multi-LLM research using configurable breadth (light/standard/exhaustive) and intensity (quick/standard/deep) with web search and multi-source synthesis.
/researchConducts multi-turn deep research on a codebase topic over 5 iterations, tracing code paths with citations, Mermaid diagrams, tables, and confidence ratings.
/researchUnderstand the opportunity and landscape: keyword demand, SERP intent, competitors, content gaps, and site/topic/entity maps. Supports --competitors and --map flags.
/researchConducts a multi-database scientific research investigation using ToolUniverse, calling tools to look up specific claims, cross-validate, and report honest INDETERMINATE results when evidence is insufficient.
/researchGathers external knowledge on a given topic, citing sources so builds stay grounded in facts. Spawns sub-agents for broad sweeps, distills findings into cited rows.