Inchin
Community@zhuyingqin
CINVESTAV-IPN (National Polytechnic Institute)
Agent Skills by Inchin
Showing 82 vetted skills indexed across 3 GitHub repositories.
zhuyingqininterview-cheatsheet
Generate long-form ML/LLM interview cheat sheets with math derivations and PyTorch code.
zhuyingqinwiki-enrich
Generate AI summaries to fill research wiki paper TODO sections.
zhuyingqinrender-html
Convert ARIS Markdown/JSON artifacts to HTML with templates and sanitization.
zhuyingqinoutlook-email
Automate Outlook mail tasks including extraction, cleanup, drafting, and routing.
zhuyingqingmail
Manages inboxes by searching, filtering, and summarizing email threads for users.
zhuyingqinopenalex-search
Automate reproducible literature search queries in the OpenAlex database.
zhuyingqincomm-lit-review
Search IEEE Xplore and ScienceDirect for communications literature and output structured reviews.
zhuyingqinproof-checker
Checks and repairs LaTeX mathematical proofs via adversarial review and audit.
zhuyingqinpaper-talk
Generate conference talk packages with Beamer slides and PPTX from academic papers.
zhuyingqinscopus-search
Execute Scopus boolean queries via elsapy and export normalized literature records as JSON.
zhuyingqinpaper-write
Generate venue-formatted LaTeX paper drafts from outlines and narratives.
zhuyingqinpaper-poster
Generate print-ready A0/A1 conference posters from compiled research papers.
zhuyingqinwriting-systems-papers
Plan paragraph-level structures for 10β12 page systems conference papers.
zhuyingqinqzcli
Manage GPU compute jobs on the Qizhi platform via kubectl-style qzcli commands.
zhuyingqinspecification-writing
Generate complete patent specifications from claims and invention disclosures.
zhuyingqinproof-writer
Write rigorous mathematical proofs for ML/AI theory claims with dependency maps.
zhuyingqinpaper-claim-audit
Cross-reference quantitative paper claims against raw evidence files and emit audit verdicts.
zhuyingqinresubmit-pipeline
Orchestrate a text-only paper resubmission pipeline with compile and citation auditing.
zhuyingqinserverless-modal
Translate GPU tasks into Modal launchers for training, inference, or batch processing.
zhuyingqinformula-derivation
Structure scattered equations and assumptions into a coherent derivation package.
zhuyingqintraining-check
Detect training divergence and quality regressions from WandB metrics.
zhuyingqinmeta-optimize
Analyze ARIS usage logs to propose evidence-backed SKILL.md optimizations.
zhuyingqinpatent-novelty-check
Map invention claim elements to prior art and analyze anticipation and obviousness.
zhuyingqinvast-gpu
Provision and manage vast.ai GPU instances for training tasks.