RigorPilot Research Skills
Auditable deep learning research with reproducibility and scientific rigor
What You Can Do
RigorPilot is a comprehensive skill suite for deep learning research, providing auditable repository analysis, README-first reproducibility workflows, bounded exploration with evidence-based ranking, and publication-ready scientific documentation. Each skill enforces rigor through documented assumptions, comparability tracking, and explicit decision checkpoints. Use these skills to understand ML repository internals, faithfully reproduce published results, systematically explore novel ideas with proper baselines, and generate standardized research artifacts.
Features
Read-only deep inspection of deep learning repositories, mapping model architectures, training/inference entrypoints, configs, insertion points, and flagging suspicious implementation patterns without modification.
Faithful reproduction prioritizing README guidance, selecting minimal trustworthy targets (inference before evaluation before training), and recording all deviations, assumptions, and environment state.
Bounded experimental work on frozen branches or checkpoints. Gates ideas with explicit scoring, runs smoke tests, and ranks candidates by real evidence before committing to full execution.
Automatic setup of datasets, checkpoints, dependencies, and caches required for selected reproduction or exploration targets without silent protocol changes.
Generates standardized outputs including SUMMARY.md, SCIENTIFIC_CHANGELOG.md, COMPARABILITY_REPORT.md, ANNOTATED_README.md, and machine-readable status.json for publication-quality evidence.
Records all executed commands, environment variables, code modifications, human decision points, and reproducibility assumptions with conservative patch governance rules.
Loads contextual principles for research safety, deep learning experiments, comparability boundaries, and patch impact to ensure decisions meet scientific standards and align with repository intent.
Example Output
Example 1: Reproducing Baseline Results
I want to reproduce the baseline inference results from the README. What's the smallest target and what do I need to set up?
Claude reads the README, extracts documented inference commands, bootstraps required datasets and checkpoints, executes the minimal target with full logging, and produces SUMMARY.md, COMMANDS.md, SCIENTIFIC_CHANGELOG.md, and COMPARABILITY_REPORT.md showing exact environment, deviations, and metrics reproduction.
Example 2: Exploring Architectural Variants
I have a frozen current_research branch with baseline results. Can you propose and smoke-test 3 architectural variants and rank them by expected impact?
Claude analyzes the repository, gates candidate ideas against the frozen SOTA reference and evaluation method, ranks each by cost and success likelihood, executes one as a smoke test, collects evidence, and documents all decisions with rollback instructions in explore_outputs/.
What's Included
- ai-research-reproduction skill: End-to-end workflow for README-first trusted reproduction with minimal target selection, auditable execution, and standardized documentation
- ai-research-explore skill: Bounded candidate exploration with idea gating, evidence ranking, and scientific comparability tracking for novelty assessment
- analyze-project skill: Read-only repository analysis for architecture, entrypoints, config relationships, and suspicious patterns
- Research governance policies: Embedded references for research rigor, deep learning experiments, patch safety, and continuous learning guidance
- Standardized artifact templates: Pre-configured markdown and JSON formats for SUMMARY.md, SCIENTIFIC_CHANGELOG.md, COMPARABILITY_REPORT.md, and status.json
- Orchestration helpers: Python utilities and scripts for artifact writing, experiment orchestration, and lesson store integration
Who It's For
- Deep Learning Researchers
- ML Paper Authors
- Research Engineers
- Benchmark and Evaluation Specialists
Best For
- Reproducible research workflows
- Auditable experiment exploration
- Scientific rigor and documentation
- Repository-grounded ML analysis
- Publication-ready evidence generation
You might also like
This comprehensive collection includes 79 dedicated agent skills for all major Azure services and operations. You can automate Azure resource deployment and management, query infrastructure and logs with natural language using Kusto Graph and IRQL, analyze security events and detect threats, configure identity and access management through Entra ID, monitor application performance with Application Insights, manage storage and messaging services, execute cloud migration workflows, validate compliance policies, and query Azure resources by metadata or properties.
Search the web, scrape URLs into markdown/HTML/screenshots/links/images, extract structured JSON with AI prompts, crawl multi-page sites, and monitor pages for changes on a schedule. Supports browser automation features like JavaScript rendering, stealth mode, cookies, and custom headers. Perfect for data engineers, researchers, and developers who need API-free web data collection without writing scrapers.
This comprehensive skill collection guides you through creating professional videos programmatically using Remotion and React. Create new video projects, structure complex multi-scene compositions, and animate content with full interactivity. Add captions by transcribing audio, importing SRT files, or generating them programmatically. Build animated maps with Mapbox, MapLibre, or GeoJSON, including 3D geographic flyovers. Process multimedia by trimming, cropping, and extracting metadata from video and audio files.
Render dynamic videos directly from HTML/code using Puppeteer and ffmpeg, with built-in support for AI voiceover, caption overlays, and motion animation. Convert markdown changelogs into polished 45-60 second branded videos with animated UI mockups, captions, and AI narration. Create explainer videos, product demos, and animated visualizations using code-first workflows and motion doctrines that enforce consistency and quality.
by heygen-com
This skill generates customized countdown timelines tailored to your launch type, complete with pre-assigned stakeholder handoffs and clear approval gates. You get built-in risk matrices that surface potential delays before they happen, plus go-no-go decision frameworks that keep your team aligned. Launches execute on schedule, cross-team confusion disappears, and you ship with confidence.
Create production-ready mobile app designs by describing what you want in plain language. Use chat-based requests to design screens, iterate with follow-up messages, and export code implementations for HTML, React Native, and SwiftUI. Features include live design preview in the Sleek editor, component screenshot generation, and built-in design references to seed your visual style.
Present users with an adaptive 24-question personality assessment that uses Bayesian active learning to efficiently map their psychological profile. The skill automatically maps user responses (1-7 scales or natural language) to archetype classifications based on a 5-color psychological model, tracks probability distributions in real-time, and delivers detailed personality insights including core strengths, weaknesses, and top archetype matches.
This collection of skills improves code quality, architecture documentation, and team workflow coordination. Review Architecture Decision Records with fresh perspective and rewrite them for clarity, implement type-safe AST visitor patterns for robust code organization, automate Biome linting upgrades across monorepos, and generate well-structured GitHub pull requests integrated with Linear project tracking. Perfect for teams that care about maintainable code, clear decision documentation, and smooth development workflows.