Labnb Promote
Promote a labnb idea into a concrete experiment with explicit budgets, source links, and provenance.
Orcd Remote
Use MIT ORCD (Engaging, orcd-login.mit.edu, eofe7) as a remote execution environment, driven over SSH from your laptop or workstation - not from the cluster itself. Use this skill whenever the user mentions ORCD, Engaging, "the MIT cluster", Slurm, sbatch/srun/squeue, running or training on cluster GPUs (H100/H200/A100/L40S), cluster scratch or storage or quotas, ssh problems reaching orcd-login,…
Labnb Idea
Record a promising but not-yet-implemented experiment idea in the global lab notebook index.
Labnb Run
Create and run a concrete lab notebook experiment with isolated workspace, explicit budgets, and iterative logging.
Kya
Govern and review autonomous agents with veldt-kya (Know Your Agents) — risk-score an agent, run multi-judge consensus, detect configuration drift, and emit compliance evidence — then turn the result into a verdict that labnb's loop can act on. Use when a run needs a trust/authorization/policy check, not just a resource or metric check.
Labnb Resume
Summarize prior ideas and experiments for a project slug, then choose whether to resume, promote, branch, or start new work.
Duct
Wrap any command with con/duct to capture wall-clock time, CPU, and memory usage as structured logs, so agents and reviewers can inspect what a run actually consumed instead of guessing.
Brainkb
>-
Labnb
Create and maintain a concurrency-safe global lab notebook outside project roots, with idea capture, startup summaries of prior work, append-only indexing, and isolated experiment workspaces across projects, investigations, and tasks.
Synthscholar
Set up, run, and query SynthScholar / PRISMA systematic reviews. (1) Protocol intake — draft a complete protocol (research question, PICO, eligibility, databases, risk-of-bias tool, charting, per-group analysis, registration) and validate it before a run. (2) Provenance queries over a finished review — full-text vs abstract-only inclusions, retrieval routes, and the screening audit trail (decisio…
Structsense
Extract structured information (named entities, key terms, resources like tools/datasets/models/benchmarks, or any target JSON schema) from unstructured text and PDFs using a model-agnostic multi-stage pipeline (extract → align to ontologies → judge → optional human feedback). Use this skill when the user asks to do NER, pull resources out of papers, convert documents to a target JSON schema (e.g…