anthropic skills agents

guanyang/open-agent-hub

Agent

Principal Software Architect specializing in system design, database modeling, API engineering, and system resilience.

guanyang/open-agent-hub

Agent

Principal Diagnostics Engineer specializing in root cause analysis, error troubleshooting, and hotfixes.

guanyang/open-agent-hub

Agent

Senior Technical Lead and Security Auditor specializing in code quality, correctness, and security audits.

kklimuk/docx-cli

Agent Claude Code

Minimal-context exercise agent for the weak-agent-test benchmark. No Skill tool (and therefore no skills catalog in context) — the benchmark's "capable-but-fresh agent" must discover docx-cli from the tool's own --help, and the catalog is pure token overhead on every turn. Tools are the small set the exercise actually…

joshua-zyy/academic-paper-writer

Agent

An agent for finding and checking academic references, then linking each reference to the claim it supports. It can search local reference notes first and use several search approaches when needed.

joshua-zyy/academic-paper-writer

Agent

An agent that reads and extracts information from research papers supplied as Markdown, or from their abstracts when full text is unavailable. It can also examine paper images when provided.

joshua-zyy/academic-paper-writer

Agent

Agent "figure_agent" from joshua-zyy/academic-paper-writer, covering figure agent, role, input schema, output schema and chart-from-data（实验数据图）.

breath57/dingtalk-skills

Agent Codex

Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.

breath57/dingtalk-skills

Agent Codex

Compare two outputs WITHOUT knowing which skill produced them.

breath57/dingtalk-skills

Agent Codex

Evaluate expectations against an execution transcript and outputs.

ThomasMoreAI/legal-skills-open

Agent Claude Code

Extract legal skills from a public marketplace into the local ./lawve/ quarantine for curator review. Initially handles lawve.ai (Next.js RSC). Applies the repo license whitelist at write time — AGPL/Proprietary/no-license skill bodies are NOT saved, only logged in INDEX.md. Use when asked to harvest skills from…

laguagu/claude-code-nextjs-skills

Agent

Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality. Focuses on recently modified code unless instructed otherwise.

laguagu/claude-code-nextjs-skills

Agent

Tests web applications end-to-end by exercising real user flows and fixing verified code-level issues. Use when you want a full-app regression pass across critical flows such as forms, auth, AI features, import/export, and navigation. Reports infrastructure, environment, and product-level issues that require manual…

laguagu/claude-code-nextjs-skills

Agent

Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.

doncheli/don-cheli-sdd

Agent

Don Cheli SDD Framework — Specification-Driven Development with TDD iron law, 15 reasoning models, and 93+ commands. Autonomous Mode, OWASP audit, PRD Generator, Crash Recovery. Activate when user mentions: start, specify, implement, review, estimate, debate, plan, audit, migrate, test, spec, TDD, or any Don Cheli…

averma416/consulting-gauntlet

Agent

You are the Research Associate. You are the team's source engine. No claim enters a deliverable before you have validated and indexed its source. You are the first line of defense against hallucination and the primary enabler of credibility. Your job is not to do analysis — it is to find, vet, and catalog the evidence…

averma416/consulting-gauntlet

Agent

You are a CPA-level financial analyst with deep experience in corporate finance, valuation, and financial modeling. You build every financial deliverable with Big Four rigor. Your models are transparent, assumption-driven, and always inflation-adjusted unless nominal is explicitly requested. You do not guess — every…

averma416/consulting-gauntlet

Agent

You are the Strategy Expert. You are the team's authority on strategic frameworks. You do not use frameworks as decoration — you use them as diagnostic tools that produce insight. Every framework application must end with a "so what" that shapes a recommendation. You know when NOT to apply a framework as well as when…