ReddRadar
Agent skills
Skills you can use with AI coding agents, indexed from public GitHub repositories.
-
metrics-tree
Use when setting product North Star metrics, decomposing high-level business metrics into actionable sub-metrics and leading indicators, mapping strategy to measurable outcomes, identifying which metrics to move through experimentation, understanding causal relationships between metrics (leading vs lagging), prioritizing metric improvement opportunities, or when user mentions metric tree, metric decomposition, North Star metric, leading indicators, KPI breakdown, metric drivers, or how metrics connect.
majiayu000/claude-skill-registry 163
-
stickies
Display and read Stickies notes on macOS. Use when the user asks to show notes, create stickies, or read existing stickies.
majiayu000/claude-skill-registry 163
-
evryg-kb:article_review
Review French articles and translate them to English. Use when the user asks to review, check, or translate KB articles.
majiayu000/claude-skill-registry 163
-
laravel-multi-tenancy
Multi-tenant application architecture patterns. Use when working with multi-tenant systems, tenant isolation, or when user mentions multi-tenancy, tenants, tenant scoping, tenant isolation, multi-tenant.
majiayu000/claude-skill-registry 163
-
person-profiler
Perform preliminary internet research on a person and generate a structured report. Use when the user asks to research, profile, or get information about a specific person by name.
majiayu000/claude-skill-registry 163
-
artifact-collector
Use this skill when users need to collect, manage, or analyze forensic artifacts such as files, memory dumps, Windows Event Logs, Mac Unified Logs, or network packet captures (PCAP) from endpoints.
majiayu000/claude-skill-registry 163
-
accessibility-issues
Use when animation excludes users with vestibular disorders, cognitive disabilities, or assistive technology needs
majiayu000/claude-skill-registry 163
-
time-resource-estimator
Estimate realistic project timelines, break down hours by feature/phase, include buffer for revisions, set achievable client expectations, and avoid over-promising deadlines you can't meet.
majiayu000/claude-skill-registry 163
-
proof-architect
Structured methodology for constructing and verifying mathematical proofs in statistical research
majiayu000/claude-skill-registry 163
-
trace-production-issue
Trace production alerts and issues back through REQ-* to original intent, creating new intent for remediation. Closes feedback loop from Runtime → Intent. Use when production alerts fire or issues discovered.
majiayu000/claude-skill-registry 163
-
plan-prd
Create product requirement documents when user wants to plan features, write specs, or document new functionality. Supports minimal core features, focused expansions, and task-based changes. Automatically loads context from related PRDs to maintain consistency.
majiayu000/claude-skill-registry 163
-
flow-designer
Design features through brainstorming and make architectural decisions. Use when user wants to brainstorm, design approach, make architecture choices, or update PLAN.md.
majiayu000/claude-skill-registry 163
-
Error Handling Gate
Verify errors are handled gracefully with meaningful user feedback. Issues result in WARNINGS.
majiayu000/claude-skill-registry 163
-
mechinterp-state
Manage research state for SAE feature analysis including hypotheses, evidence tracking, and research history
majiayu000/claude-skill-registry 163
-
jira-safe
Implement SAFe methodology in Jira. Use when creating Epics, Features, Stories with proper hierarchy, acceptance criteria, and parent-child linking.
majiayu000/claude-skill-registry 163
-
forms-inputs
Use when animating form fields, inputs, textareas, selects, or interactive form elements to improve usability and feedback
majiayu000/claude-skill-registry 163
-
jj
Use the jj version control system. Always use it when the user mentions jj.
majiayu000/claude-skill-registry 163
-
product-self-knowledge
Authoritative reference for Anthropic products. Use when users ask about product capabilities, access, installation, pricing, limits, or features. Provides source-backed answers to prevent hallucinations about Claude.ai, Claude Code, and Claude API.
majiayu000/claude-skill-registry 163
-
research-explorer
Conducts external research for competitive analysis, standards, and APIs
majiayu000/claude-skill-registry 163
-
Pathfinding
Collaborative Q&A to clarify unclear requirements. Use when user wants to figure something out, brainstorm ideas, work through a challenge, explore approaches, or plan features. Triggers would be when requirements are unclear, when exploring an idea, when planning, or when a user explicitly mentions pathfind(ing), brainstorm(ing), clarify(ing), figur(ing) out, work(ing) through, or `--pathfind`.
majiayu000/claude-skill-registry 163
-
qiita_team_get_item
社内Qiita:Teamのドキュメントコンテンツを取得する。コード内のコメントやドキュメントにQiita:TeamのURL（https://[team_id].qiita.com/[user_id]/items/[item_id]形式）が含まれている場合に使用する
majiayu000/claude-skill-registry 163
-
Checking theme alignment
Validates that book chapters align with the core theme. Use when user says 'check theme alignment', 'does this fit the theme', 'verify alignment', or automatically after every 2 chapters are drafted.
majiayu000/claude-skill-registry 163
-
clarification-expert
Clarify ambiguous requests by researching first, then asking only judgment calls; stop before implementation.
majiayu000/claude-skill-registry 163
-
shell-tools
Production-grade shell tools - jq, xargs, parallel, pipelines
majiayu000/claude-skill-registry 163