bingran-you

A minimal, opinionated GitHub label taxonomy for OSS / internal projects covering priority, area, issue status, PR review state, and independent reproduction. Use when setting up labels for a new repo, when triaging a backlog, when asked "how should we label issues", when reviewing whether existing labels are coherent, or when applying labels to a batch of open issues. Five orthogonal axes, ~16 labels total, every label answers a specific filter query — designed against the open-source convention of `S-waiting-on-*` (Rust) and two-stage approval (Kubernetes), but kept small enough for a solo / small-team repo to actually maintain. Includes a bootstrap script (`scripts/bootstrap-labels.sh`) that creates the full label set in a target GitHub repo with one `gh` call per label.

Run an extremely strict maintainability review for abstraction quality, giant files, and spaghetti-condition growth. Use for a thermo-nuclear code quality review, thermonuclear review, deep code quality audit, or especially harsh maintainability review.

Manage Overleaf-backed LaTeX papers mirrored to GitHub with bidirectional GitHub Actions sync. Use for Overleaf git tokens, paper version control, GitHub mirrors, divergence or merge conflicts between Overleaf and Git, repos with sync-overleaf/pull-from-overleaf workflows or sync helpers, collaborator onboarding, and named paper instances listed in references/instances.md.

REQUIRED before reading, scraping, navigating, or extracting metadata from X / Twitter, Xiaohongshu / RedNote, YouTube, Bilibili, or LinkedIn, including the bingran.ai /posts pipeline. Trigger for social URLs, "add this to /posts", "go look at X", "grab those posts", mention summaries, social heartbeats, or any fetch/browser automation touching those domains. Defines safe account, tool, pacing, fingerprint, abort-signal, data-handling, and post-card rules.

Apply the bingranyou.com Win98 paper site-wide theme to a Next.js + Tailwind site so non-palace routes share the same visual language as /palace. Use when porting palace typography, palette, CJK fallback, paper-smudge overlay, or raised Win98 CTA styling to the front door, About, blog, or other routes.

Reproducible kit for mounting Henry Heffernan's 3D-CRT-monitor + Win98-desktop portfolio template (`/palace`) on top of an existing Next.js site. Ships drop-in build scripts, config snippets (webpack / Next / Vercel / tsconfig), placeholder showcase pages, NOTICE templates, a post-deploy DevTools verification snippet, and the 13-gotcha PLAYBOOK that catalogues every way this rollout breaks in production. Use when the user wants a /palace-style 3D portfolio with a clickable retro monitor mounted on their own site, when integrating a vendored CRA app inside a Next.js host, or when debugging "same-origin iframe but Chrome treats it as cross-origin", iframe stuck on broken-document, framer-motion type drift on Vercel, COOP/X-Frame-Options blocking own-iframe, or font fallback inside the inner OS.

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

Deep web research benchmark — given partial clues about authors and topic, find the exact paper and produce a structured answer file with the paper title and DOIs. Use when the user wants to test or demonstrate an agent's ability to do multi-hop, citation-aware web research from sparse input. Originally authored as a SkillsBench task by Bingran You; copied here as a discoverable, self-contained reference.

Equity research benchmark — produce a one-page US-stock investment memo from a fixed local data snapshot, with mandatory chapter structure (Decision Card, Dual-Horizon Framing, Verified Facts, Derived Metrics, three-tier scenarios, Triggers, Judgment), arguing from both prospective-buyer and existing-holder viewpoints. Use when evaluating an agent's structured financial-analysis output. Originally authored as a SkillsBench task by Bingran You; copied here as a discoverable, self-contained reference.

Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.

Create stunning, animation-rich HTML presentations from scratch or by converting PowerPoint files. Use when the user wants to build a presentation, convert a PPT/PPTX to web, or create slides for a talk/pitch. Helps non-designers discover their aesthetic through visual exploration rather than abstract choices.

Install Codex skills into $CODEX_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).

Use when writing, planning, analyzing, or editing Xiaohongshu (小红书) content — posts, titles, covers, hashtags, captions, account positioning, growth strategy. Provides on-demand access to a 139-skill knowledge base covering content creation, account ops, interaction, analytics, e-commerce, platform rules, tooling, marketing, and growth. Read this index first to locate the relevant sub-skill, then read that sub-skill's SKILL.md before producing output.

小红书相关操作，覆盖账号定位、选题研究、内容生产、发布执行与复盘修复的小红书全链路运营技能。凡是小红书的浏览/搜索/发布/评论任务，默认必须使用 OpenClaw 内置浏览器流程并指定 profile="openclaw"；除非用户明确要求，否则不要使用系统 open 或外部浏览器。