Designs multi-agent architectures from requirements, generates validated Anthropic/OpenAI tool schemas, and evaluates execution logs for cost, latency, and bottlenecks.
Compiles a goal into a verifiable task plan and enforces an execute–verify–retry–escalate loop through a state machine.
An Arabic-language planning skill that splits complex work into task_plan, findings and progress Markdown files and auto-restores the plan after context is cleared.
Manus-style persistent planning that keeps task_plan.md, findings.md and progress.md on disk so multi-step work survives compaction and /clear.
A German-language skill that plans and tracks complex, multi-step work in task_plan.md, findings.md and progress.md, and restores context after /clear.
Audits and rewrites content so ChatGPT, Perplexity, Claude, Gemini, and Mistral cite it as the authoritative source.
Manus-style persistent planning that keeps task_plan.md, findings.md, and progress.md on disk so work survives /clear and context loss.
Manus-style file-based planning that tracks complex work in task_plan.md, findings.md and progress.md — and restores context after /clear.
Designs statistically valid A/B tests end to end — hypothesis, sample size, guardrails, and result analysis.
Generates, iterates, and scales paid ad copy that fits Google, Meta, LinkedIn, X, and TikTok character limits.
Keeps task_plan.md, findings.md and progress.md on disk so multi-step work survives context loss and /clear.
A reference skill for designing, submitting, and retrieving protein experiments through the Adaptyv Foundry API and Python SDK.
Teaches Claude how to create, read, concatenate and optimize AnnData objects (.h5ad/.zarr) in the scverse ecosystem.
Guides Claude through classification, regression, clustering, forecasting, and anomaly detection on time series using the aeon toolkit.
Query 40+ bioinformatics databases such as UniProt, KEGG, ChEMBL and Reactome through one consistent Python interface.
Helps organize, query, validate, and convert neuroscience data (MRI, EEG, PET and more) into the BIDS standard.
Helps Claude write and debug Astropy code for units, coordinates, FITS I/O, tables, time scales, WCS and cosmology.
Analyzes your local screenpipe history to detect repeated research workflows and drafts new skills or composition recipes for whatever isn't covered yet.
Orchestrates a reproducible bulk RNA-seq workflow from raw FASTQ through QC, quantification, differential expression, pathway enrichment and publication figures.
Runs a long-horizon experiment loop that improves a scorable artifact through a persistent hypothesis tree, with a held-out test gate to block overfitting.