Research

Default entry point for any research request — a hybrid router that classifies the question deterministically and either delegates to a specialist research…

Rodado's prospector.

Entry point for experiments whose printed numbers must be traceable: freezing corpus, samples, gold sets and model or prompt versions, the runs log, numbers…

Draws explanatory figures for learning: scene cards, four-panel comics, textbook and auto-laid-out DAGs, mind maps, explainer pages and steppers for proofs…

Draft and check a prospective experiment contract from a research question, comparison, estimand, experimental units, control, falsifier, harm limit and…

Research a topic to support a decision, three ways: draft a research prompt for the user to run in their own tool (ChatGPT, Gemini, Grok, Perplexity, …), turn…

Run a multi-question investigation and produce a durable, multi-page research report that lives next to the project: clarify scope, gather evidence with…

On Claude Fable 5.1 or Opus, subagents inherit the session's expensive model and can spawn more of themselves, so one research fan-out can burn the usage…

GROUND leg: assesses ACL/citation/refusal probe evidence into a threadlight.ground/v1 manifest (GRD-001..004: ACL, citation, refusal…

Use when the user asks to research something on the web, search, crawl or map a site, or pull content from URLs.

Deep multi-source, fact-checked research report via the Workflow harness.

Project registry — create, list, and look up first-class projects (multi-week research work bound to a doc, chat, and schedule); read/write the canonical…

Create and improve source-grounded scientific SVG diagrams and animated worked examples.

Compares source-linked table-cell and figure-caption records from old scientific PDFs with a separate adjudicated reference, counting abstentions and…

Use when the user asks to generate daily paper digests on a general topic.

Render a single "where am I, what have I done, what's next" survey HTML page composing anchor cascade + tesseract shelf + git state + CHECKPOINT.md + TaskList.

Ground answers and documents in cited, verifiable sources.

Use ARIS, the Auto Research in Sleep skill-based research harness, for setup, workflow routing, reviewer backend integration, state recovery, experiment…

Operate the Deep Researcher Agent repository for autonomous deep-learning experiments, execution backends, provider/tool dispatch, GPU resources, durable…

Routes Autograd install, differentiation, wrapper, extension, and optimization workflows for NumPy-based scientific Python code.

Guides DeepMotionEditing BVH animation, skeleton-aware motion retargeting, motion style transfer, and Blender visualization workflows for research use.

Operate Tongyi DeepResearch and its WebAgent family: ReAct inference setup, rollout validation, benchmark evaluation, and variant routing.

Use DeepXDE for scientific machine learning, PINNs, DeepONet/operator learning, backend selection, training, and troubleshooting.

Use the DeepXiv SDK for citation-aware academic and web research through its Python Reader, deepxiv CLI, progressive paper APIs, optional local LangGraph…