yakeworld
Community@yakeworld
yakeworld maintains Synthos, an MIT-licensed cognitive-atom skill system for academic research pipelines: literature retrieval, knowledge extraction, hypothesis generation, and quality-gated paper production.
Agent Skills by yakeworld
Showing 149 vetted skills indexed across 1 GitHub repositories.
yakeworldviewpoint-verification
Validates hypotheses and arguments through falsification testing, Bayesian scoring, and robustness analysis.
yakeworldtask-router
Routes user queries to cognitive atom chains and records execution in pipeline_trace.json.
yakeworldquality-gate
Validates research papers through layered quality gates with automated runner scripts and repair loops.
yakeworldknowledge-extraction
Extract structured entities, relations, claims, and evidence from academic papers into KnowledgeItem JSON.
yakeworldknowledge-acquisition
Search academic literature and download verified PDFs via DOI with multi-source fallback.
yakeworldhypothesis-generation
Generates falsifiable, scored research hypotheses from gap analysis and association outputs.
yakeworldassociation-discovery
Identify cross-paper associations and research gaps from structured knowledge items.
yakeworldargument-expression
Transforms hypotheses into structured academic arguments using Toulmin, CARS, and IMRaD frameworks.
yakeworldresearch
Monitors PubMed and arXiv literature across multiple research directions with local date filtering.
yakeworldk230-development-lessons
Automate K230 board management via direct Pyboard serial calls instead of ampy run.
yakeworldxhs-content
Generates Xiaohongshu post packages with body text, tags, cover copy, and compliant figures.
yakeworldreference-directory-cleanup
Reconciles PDF files in paper directories against BibTeX entries and verifies LaTeX compilation.
yakeworldpaper-improvement-patterns
Improves compilable LaTeX papers by adding frontier citations, ablation studies, statistical tests, and availability statements.
yakeworldgoogle-search
Execute web searches through a four-level fallback chain of APIs, SearXNG, and Tor.
yakeworldpidd-leakage-audit
Detects and fixes data leakage in PIDD machine learning pipelines using fold-isolated preprocessing.
yakeworldmlops
Route machine learning operations requests to subdomain skills for training, inference, evaluation, and deployment.
yakeworldlatex-compilation-troubleshooting
Diagnose and fix LaTeX compilation errors, BibTeX citation failures, and elsarticle package conflicts.
yakeworldpima-audit
Audit machine learning experiments for data leakage and validate claims against published literature.
yakeworldpaperjury
Reviews research papers by independently reproducing experimental metrics and auditing claims against code outputs.
yakeworldcompetition-proposal
Fill DOCX competition application templates with structured four-part narratives using python-docx.
yakeworldellipse-to-3d-circle
Reconstructs 3D circle geometry from 2D ellipse projections using camera intrinsics and prior constraints.
yakeworldpaper-experiment-audit
Audits LaTeX paper experiments for dataset version consistency, ablation reproducibility, and citation integrity.
yakeworldlinux-environment-troubleshoot
Diagnose and repair Linux environment failures including MiKTeX locks, Python venv isolation, and dpkg interruptions.
yakeworldsynthos-skill-audit
Audit AI skill files against the Synthos five-section standard with privacy scanning.
Frequently Asked Questions About yakeworld
FAQPage SchemaWhat tasks can I automate using yakeworld's Synthos skills?▼
You can automate literature search and PDF download, structured knowledge extraction from papers, cross-paper association discovery (contradiction, complement, evolution, gaps), hypothesis generation, citation verification with DOI back-checking, and quality-gated LaTeX paper production with fixed pipeline gates.
Who is the target user for these skills?▼
Graduate students, academic researchers, and research engineers running paper-production pipelines. Supporting skills also serve DevOps operators (cron diagnostics, Linux environment repair, privacy scanning) and content creators (XHS posts, figures, presentations), but the core audience is clinical and ML research authors.
How do the skills execute at runtime?▼
The task-router skill analyzes a query, selects an execution mode, builds an atom chain, and dispatches cognitive or mechanical atoms, emitting a pipeline_trace.json. Skills run via terminal, file read/write, web search, and session search permissions, with chained pipelines like ACQ→EXT→download→quality-gate.
Are yakeworld's skills open source and what do they cost?▼
Yes. Every skill in the manifest carries an MIT license, so they are free to use, modify, and redistribute. There is no stated cost; some skills depend on external services (Semantic Scholar, Crossref, PubMed, model endpoints) that may have their own terms or API keys.
What prerequisites and dependencies do the skills require?▼
Skills assume a terminal environment with file read/write and session search access. Specific pipelines reference jabkit-rs, doi-fetch, pdflatex, Playwright, and Jupyter kernels; several cognitive atoms pin the anthropic/claude-4-opus-max model and enforce a 0.85 golden-set pass threshold.