Skills

Agent Skills12 Claude Code9 Codex5 Design System3 Frontend3 TDD3 Academic Writing2 Code Review2 Cursor2 Deep Research2 HTML Slides2 Presentation Design2 Progressive Disclosure2 SVG2 Software Engineering2 UI Design2

ResearchzLanqing/codex-claude-academic-skills

Three Chinese-first research skills that run on both Claude Code and Codex

Editor's takeThe part worth copying is that academic integrity is written as instructions: no fabricated data, DOIs or journals, and every piece of content is classified as quotation, confirmed fact, reasonable inference or writing suggestion, so the author can see at a glance which sentences have a source. The real danger in academic writing is not weak prose but a model writing inference as though it were established fact, and that four-way split blocks exactly that. The three skills also compose: figures, then body text, then defence slides.

4,117 stars, MIT. research-writing handles drafting, revising and polishing paper sections plus rebuttals; office-academic generates and edits academic Word and PPT at the OOXML layer (literature reports, group meetings, proposals, defences); scientific-toolkit covers MATLAB/Python simulation, signal processing, statistics, machine learning and paper figures, with literature search across arXiv, PubMed, CrossRef, Semantic Scholar and OpenAlex.

Claude CodeCodex

git clone https://github.com/zLanqing/codex-claude-academic-skills && cp -r codex-claude-academic-skills/research-writing-skill ~/.claude/skills/

PythonzLanqing4.1k

Developmentunclebob/ub-mode

ub-mode: A Discipline Layer for Coding Agents

Editor's takeThe discipline is compact, but its quality gates depend on external tools such as Crapper, Dryer, and Mutator.

A behavior-only agent skill that imposes a three-phase engineering loop: make the change work, then make it right, then keep the process auditable with logs, context control, and small commits.

Claude CodeCursor

Copy SKILL.md into ~/.claude/skills/ub-mode/SKILL.md, then invoke /ub-mode

unclebob4k

Docs & SlidesGordenSun/GordenPPTSkill

Gorden PPT Skill: 21 Chinese Templates and a Non-Destructive Contract for Real .pptx Decks

Editor's takeA rare PPT skill that writes taste as decidable clauses: soft capacity, banned ellipses, consistent type scale, all aimed at real agent failure modes.

GordenSun Gorden PPT Skill (v1.0.20, 3,748 stars) turns "make me a deck" into a reviewable pipeline: 21 built-in Chinese templates each ship intro.md and detail.json, the agent selects slides by page_roles and writes edits.json, and build_pptx.py replaces run text only, never position, size or color; overflow checks warn without blocking, ellipsis truncation is banned by rule, and same-level titles must keep their type_scale size; user-supplied templates are rendered and probed on the spot, original layouts are only a fallback; manifest.json plus updates.json give the skill software-style incremental self-update. Code is MIT; templates carry a separate non-commercial NOTICE.

Any agent

PythonGordenSun3.7k

Data & Analyticstradermonty/claude-trading-skills

74 trading skills for process and risk, explicitly not for signals

Editor's takeWhat earns trust is the self-limitation: it states it is not designed for fully automated trading, signal outsourcing or scalping, a long disclaimer tops the README, and the licence is MIT 'AS IS'. The swing workflow runs drawdown-circuit-breaker and pre-trade-discipline-gate before any screener — risk gates before tickers. That ordering is the soul of the design and the opposite of most tools that lead with a list of picks.

2,867 stars, MIT, Japanese README included. 74 skills across 8 categories, 58 production and 16 beta, all governed by one canonical skills-index.yaml where each entry carries a verification matrix (data source, required or recommended, standalone or workflow_step, maturity); the README catalogue is generated from it. YAML manifests under workflows/ chain the skills into executable flows enforced by a validator.

Claude Code

git clone https://github.com/tradermonty/claude-trading-skills && cp -r claude-trading-skills/skills/backtest-expert ~/.claude/skills/

Pythontradermonty2.9k

Research199-biotechnologies/claude-deep-research-skill

An eight-phase deep research pipeline with an evidence ledger

Editor's takeThe soul is four append-only JSONL ledgers: sources, evidence, claims and run_manifest. Every factual claim must carry an inline [N] backed by evidence.jsonl, unsupported claims may not ship, and validate_report.py plus verify_citations.py check the result. The repo has no LICENSE file, which under default copyright means all rights reserved: you can read and run it locally, but get the author's permission before redistributing or using it commercially.

1,051 stars. Four intensity modes — Quick at 3 phases and 2-5 minutes, Standard at 6, Deep at 8, UltraDeep at 8+ and 20-45 minutes — running SCOPE, PLAN, RETRIEVE, TRIANGULATE, outline refinement, SYNTHESIZE, CRITIQUE, REFINE and PACKAGE, and emitting Markdown, a McKinsey-style HTML report and a PDF.

Claude Code

git clone https://github.com/199-biotechnologies/claude-deep-research-skill ~/.claude/skills/deep-research

Python199-biotechnologies1.1k

SecuritySushegaad/Claude-Skills-Governance-Risk-and-Compliance

36 compliance frameworks as individual .skill packages, with a published eval

Editor's takeMost compliance prompt packs only claim expert level; this one ships numbers — 180 test cases, 902 verifiable assertions, graded by independent agents, 89% with the skill against a 57% baseline. That gap says general models are shaky on fine-grained control mapping, which is the daily work of compliance. It carries its own disclaimer: not a substitute for licensed legal or audit opinion. Its value is making the first draft complete and correct, and a human still signs.

911 stars, MIT, v2.0.0, updated monthly. Information security covers ISO 27001, SOC 2, FedRAMP, NIST CSF, NIST SP 800-53 and CIS Controls v8; privacy covers GDPR, ISO 27701, CCPA/CPRA, LGPD and DPDPA; AI governance covers ISO 42001, NIST AI RMF and the EU AI Act; sector and country law covers HIPAA, PCI DSS, DORA, SWIFT CSP, TISAX, SOX ITGC, EU NIS2/CRA/CSRD and ITAR/EAR.

Claude Code

/plugin marketplace add Sushegaad/Claude-Skills-Governance-Risk-and-Compliance

HTMLSushegaad911

MediaYusuke710/manim-skill

Let Claude Code produce 3Blue1Brown-style explainers with Manim

Editor's takeStep three is the point: the agent owns the render-until-success loop instead of handing errors back to you. That is exactly where Manim costs humans the most, since one wrong coordinate means another ten-second re-render. The repo also writes 'avoid generic AI slop aesthetics' into its goals, treating taste as a requirement rather than luck, and keeps plan.md as a separate file so iteration does not require re-explaining context.

156 stars, MIT, site manimate.ai. Four steps: Plan (storyboard, optionally in plan mode), Code (write Manim scenes in Python), Render (keep going until every scene renders successfully) and Iterate (open the video viewer in a browser and paste your notes back). The project layout is fixed as plan.md, script.py, concat.txt and final.mp4.

Claude Code

/plugin marketplace add Yusuke710/manim-skill

HTMLYusuke710156

As an Amazon Associate, we earn from qualifying purchases.