All skills

Forces a probe → hypothesis → smoke → control → claim discipline when an ML training run fails, diverges, OOMs, hangs, or looks slow.

Runs alpha/beta reader feedback as a disciplined round: collect per-reader notes, sort convergent/divergent/single findings, and output a synthesis with a readiness verdict and revision plan.

Runs the human-in-the-loop pre-publication workflows: sensitivity reads, real-people/defamation flagging, quote permissions, AI disclosure, editor rounds, and co-author collaboration.

Builds a single portfolio-level dashboard of client health, budget pacing, deadlines and team utilization across every brand you manage.

A decision-tree-driven skill for labeling cell types in scRNA-seq data via manual markers, automated classifiers, or reference label transfer.

Writes and validates Python protocols for Opentrons OT-2/Flex robots — pipetting, serial dilutions, PCR, and hardware module control.

A repo-local scaffolder that generates a schema-valid SKILL.md, registers it in registry.yaml, and runs validation for the SciAgent-Skills repository.

An umbrella design skill with six modes: UX copy, design systems, design critique, WCAG accessibility audits, developer handoff specs, and user research synthesis.

Finds GitHub Actions workflows whose latest run failed on the staging branch, summarizes the error, and hands off the fix.

Turns Claude into a Polish-law contract assistant for analysing, risk-auditing and drafting B2B, IP and IT agreements (NDA, body leasing, SaaS, copyright assignment).

Turns Brand Registry SFR, Market Basket and Item Comparison exports into an actionable growth strategy report.

Turns a journal's author guidelines into two canonical profiles — a detailed one for drafting papers and a compact one for journal matching.

Measures your note vault's link graph and reports orphan rate, average degree, components, hubs, bridges and clusters — plus what each number means for retrieval.

Picks real frames from your video and generates 3:4 / 4:3 / 16:9 thumbnails for Xiaohongshu and Bilibili in one run.

Builds positioning and a narrative where the developer is the hero and a real market trend is the villain.

A contract-review skill that flags unfavorable terms in NDAs, SaaS terms, employment agreements and M&A docs, compares them to market standards, and drafts redlines.

Turns observed failures, repeated work and avoidable waits into scoped, verified process fixes — then commits only the changes it owns.

An onboarding playbook that locks in brand facts, approved claims and voice once, then saves them so every later job reads instead of asking.

Export Feishu/Lark cloud docs to Markdown, create new documents, and patch individual blocks straight from the CLI.

Turns vague visual taste into a batch of comparable design options so the user can point at what they actually want.