Menu
Create a scheduled task that can be run on demand or automatically on an interval.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Generate publication-grade ML explainer videos and carousels the way 3Blue1Brown actually builds them — in real manimGL (NOT Manim Community Edition), as a tiny domain DSL of self-arranging Mobjects choreographed into transform-driven beats where every motion carries meaning. Overlap is prevented at construction time, not policed after render. Use for: 3b1b-style ML videos, paper-figure animations, IG carousels, infographics, posters.
Write, format, and export professional academic research papers as publication-ready PDFs using reportlab. Use this skill whenever the user wants to write a research paper, preprint, white paper, literature review, position paper, or technical report — whether from scratch, from notes, or from an existing draft. Also trigger when the user asks to convert markdown/text into an academic PDF, needs SSRN/arXiv submission preparation, or wants tables, figures, and references formatted for publication. Trigger on: "write a paper", "preprint", "format as research paper", "turn my notes/draft into a paper", "academic PDF", "SSRN", "arXiv", "research report", "literature synthesis", "position paper", "white paper", "publish my research". Even if the user just provides a markdown file and says "make this a PDF", use this skill if the content is scholarly/research-oriented.
Autonomous deep research agent that reads all files in the active directory, performs exhaustive multi-round literature research using web search and Brightdata scraping tools, iterates through self-critique loops, and produces a novel, publication-quality research paper as a formatted PDF. Trigger on: "research this", "do a literature review", "write a research paper from these files", "find what's missing in the literature", "autonomous research", "deep research", "novel research", "investigate this topic", "analyze the literature", or any request to turn seed material into a professional research output. Also trigger when the user uploads academic papers, notes, or datasets and asks Claude to "find gaps", "what's novel here", "what hasn't been studied", or "build on this".
Reflective pass over your memory files — merge duplicates, fix stale facts, prune the index.
Audit and harden existing codebases (especially AI-generated / vibe-coded ones) for production readiness. Use when the user asks to review, audit, clean up, harden, deslop, refactor, or fix quality issues across an existing codebase. Works in two phases — first a thorough multi-pass audit written to a structured file, then systematic fixes applied in safety-tiered order. Language-agnostic. Does NOT change business logic — only hardens, cleans, and robustifies.
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.
| name | schedule |
| description | Create a scheduled task that can be run on demand or automatically on an interval. |
You are creating a reusable shortcut from the current session. Follow these steps:
Review the session history to identify the core task the user performed or requested. Distill it into a single, repeatable objective.
The prompt will be used for future autonomous runs — it must be entirely self-contained. Future runs will NOT have access to this session, so never reference "the current conversation," "the above," or any ephemeral context.
Include in the description:
Write the description in second-person imperative ("Check the inbox…", "Run the test suite…"). Keep it concise but complete enough that another Claude session could execute it cold.
Pick a short, descriptive name in kebab-case (e.g. "daily-inbox-summary", "weekly-dep-audit", "format-pr-description").
Pick one:
cronExpressionfireAt ISO timestampcronExpression: Evaluated in the user's LOCAL timezone, not UTC. Use local times directly — e.g. "8am every Friday" → 0 8 * * 5.
fireAt: Compute the exact moment and emit a full ISO 8601 string with timezone offset, e.g. 2026-03-05T14:30:00-08:00. Never use cron for one-time tasks — cron has no one-shot semantics.
Finally, call the "create_scheduled_task" tool.