Menu
SkillsMP has collected 29 skills from UCL-ERL/skills. Open a skill to review its source and details.
Showing 29 of 29 collected skills.
Audits benchmark design, baselines, metrics, ablations, statistical reporting, leakage risk, and claim-result alignment for Robotics and AI papers. Use when the user asks whether experiments are convincing, fair, reproducible, or sufficient for paper claims.
Builds a portable, source-grounded Markdown dossier from a remote research project, including method state, experiment design, result ledgers, figures, failures, provenance, and paper-ready claim candidates. Use on the machine that contains the code and…
Audits whether a Robotics or AI manuscript accurately describes its implementation, configs, data pipeline, training and inference behavior, evaluation code, defaults, and generated tables or figures. Use when reconciling a paper with a codebase, preparing an…
Routes Robotics & AI coding work through design, prototyping, debugging, testing, review, and reproducibility checks. Use when the user asks to build, debug, test, review, or restructure Python, C++, ROS, simulation, ML, or experiment code.
Routes Robotics & AI conference paper writing for ICRA, RSS, CoRL, ICML, ICLR, and NeurIPS style submissions. Use when the user asks to outline, draft, revise, audit, rebut, package, or prepare a research paper for Robotics or AI venues.
Routes Robotics & AI research work across search, reading, evidence mapping, ideation, paper writing, figure planning, and audits. Use when the user asks to enter a field, build a literature base, brainstorm a project, write a survey or paper, or coordinate…
Interrogates a Robotics or AI research, paper, experiment, survey, or engineering plan one decision at a time until its objective, evidence, scope, evaluation, risks, and next artifact are explicit. Use when the user asks to be grilled, wants a plan…
Builds timed, source-grounded research talks and validated slide decks from papers, surveys, experiment dossiers, or project artifacts. Use when preparing a lab meeting, journal club, conference spotlight or oral, invited talk, project update, thesis…
Performs harsh but evidence-grounded review of Robotics & AI papers across factual accuracy, logic, novelty, experiments, citations, figures, writing, and reviewer risk. Use when the user asks for ruthless proofreading, red-team review, paper critique, or…
Audits whether citations are real, traceable, and actually support manuscript claims. Use when the user asks to verify references, check citation truthfulness, map claims to papers, detect hallucinated citations, or audit related work.
Audits limitations, assumptions, failure cases, negative results, boundary conditions, and ethical or deployment risks in Robotics and AI papers. Use when the user needs a limitations section, failure analysis, responsible research check, or pre-review…
Audits manuscript coherence, section flow, paragraph logic, repetition, contradictions, and claim threading. Use when the user asks whether a paper draft, section, abstract, introduction, related work, or conclusion is logically connected, non-redundant, and…
Turns an experiment dossier, verified literature, and venue constraints into a coherent source-grounded conference paper story and Markdown or LaTeX draft. Use when a local writing environment has research results but needs a defensible narrative,…
Edits academic paper prose for grammar, clarity, concision, scientific tone, reviewer readability, and removal of generic AI-style writing patterns while preserving technical claims. Use after claims, citations, and structure are checked, or when the user…
Builds a source-grounded Robotics or AI survey from verified reading cards, search logs, evidence matrices, and bibliography records. Use when defining a survey thesis or taxonomy, synthesizing literature across method families, designing section logic and…
Audits whether research results are traceable from paper claims back to configs, code versions, data versions, seeds, logs, hardware, and generated artifacts. Use when checking reproducibility, result provenance, experiment tracking, or paper appendix…
Onboards a new Robotics or AI research area, project folder, or lab knowledge base into a structured source inventory, concept map, artifact map, and next-action plan. Use when the user starts a new area, inherits notes/code/papers, or needs a reusable…
Builds or revises paper abstracts and introductions from verified problem, gap, method, evidence, contribution, and limitation material. Use when the user needs a conference-paper abstract, introduction, contribution paragraph, or opening narrative without…
Checks LaTeX paper source packages for venue formatting, anonymity, buildability, references, figures, arXiv readiness, and submission hygiene. Use before conference submission, arXiv upload, camera-ready, or when LaTeX build/source problems may block a paper.
Positions a paper against prior work by building source-grounded clusters, contribution boundaries, novelty claims, and comparison language. Use when writing related work, introduction positioning, survey paragraphs, or responses about novelty and…
Builds evidence-grounded, point-by-point reviewer responses and rebuttal plans for Robotics and AI papers. Use when the user has reviews, meta-review, AC comments, or rebuttal constraints and needs a concise, non-defensive response strategy.
Applies Robotics & AI code quality checks for Python, C++, ROS, simulation, ML experiments, and reproducibility. Use when implementing, debugging, testing, or reviewing Robotics and AI research code.
Plans, prompts, and audits scientific figures, diagrams, and GPT Image outputs for Robotics & AI papers. Use when the user asks where figures are needed, what figure to draw, how to prompt image generation, or whether a returned figure is scientifically valid.
Scores and refines Robotics & AI project ideas for novelty, feasibility, baselines, evaluation, risk, and publishability. Use when brainstorming, ranking, stress-testing, or improving research ideas with or without the user.
Builds comparison matrices, taxonomies, and gap maps from paper cards, projects, datasets, or benchmark notes. Use when the user asks to synthesize a literature base, write a survey, compare methods, or find research gaps.
Turns one Robotics & AI paper into a structured, source-grounded reading card. Use when the user asks to read, summarize, compare, or extract claims, methods, experiments, limitations, and reusable ideas from a paper.
Builds reproducible Robotics & AI search logs across papers, projects, datasets, benchmarks, and code. Use when the user asks for literature search, project search, survey source collection, related work discovery, or field onboarding.
Plans Robotics & AI conference paper structure for ICRA, RSS, CoRL, ICML, ICLR, and NeurIPS style submissions. Use when the user asks for an outline, paper skeleton, section plan, contribution framing, or venue-aware writing strategy.
Use this skill when an agent needs to perform one concrete, reusable workflow.