Writing & Editing

name: scrna-review-qc

Safety-first agent skill for cleaning Git worktrees tied to closed GitHub or GitLab issues

Agent skill: a ranked, summarized roundup of the latest arXiv research on AI agents (default: harness optimization) with 'read these first' picks and a saved report. Fetches via the Bright Data MCP.

An importable Paperclip Agent Skill for one bounded, authorized CAPTCHA recovery with validation and human handoff.

Agent skill for Claude Code: edits Russian text, stripping bureaucratese, AI-generation tells and cliches, with a separate spelling and punctuation pass.

Agent skill that extracts a measurable writing fingerprint from real samples and checks drafts against it so AI tells do not take over.

Agent skill that measures TPR, TNR, bias, and uncertainty so an LLM evaluator can be trusted before it scores production outputs.

Agent skill for framing ambiguous questions, keeping real alternatives open, and converging only when evidence or values support a decision.

Agent skill that captures a mid-task thought, sends bounded background work to a sub-agent, and returns the result without losing the main thread.

Writes text in your own voice instead of generic AI prose, with a check and polish pass built in. Works with Claude Code, Codex CLI, and any agent that reads skills off the filesystem.

Structural skill-library debt scan (missing-artifact detection) — the deterministic code layer of the Agent Knowledge Cycle's Curate phase.

Plan-native execution harness for AI coding agents: intent-to-contract synthesis, bounded delegation, durable handoff, and evidence-based acceptance.

Claude Code skill: extracts reusable patterns from sessions and evaluates their value before saving

Agent skill for writing scientific literature reviews with verified, retraction-screened, peer-reviewed citations only. Works with Claude, ChatGPT, and other LLM agents.

Writes AI-facing documents (llms.txt / llms-full.txt / FAQ / glossary) optimized for citation by ChatGPT, Perplexity, Gemini. Answer.AI llms.txt standard + GEO-SFE 3-layer static analysis.

Agent-first DeepSeek Harness plugin intelligence: verify existing plugins, identify missing capabilities, and generate build-ready briefs.

让 AI 像资深测试工程师一样工作：全生命周期 QA Agent Skills 框架——方法论 + 10 Skills + 可复现 Benchmark（Claude Code 等 Agent 可用）

Reusable agent skills for evidence-first technical and editorial writing

a set of claude skills for python development (and more) based on mcginniscommawill.com writings.

Codex skill for content-rich, taste-calibrated B2B brand websites with three-pass visual QA

A Codex Skill for high-impact 3:4 Chinese AI content covers with the Kasadin IP.

Codex skill for international edition nine content unit workflow

Codex skill for native Baidu Baike editing through persistent Chrome and Playwright CDP

面向信贷数据的数据质量分析 Codex Skill 与可复核的 Excel/Ollama 参考实现