By rararulab
Agent skills for development workflow, team orchestration, language learning, and requirement management
Use when designing BDD features, writing Gherkin acceptance criteria, decomposing requirements into testable scenarios, creating GitHub issues with .feature specs, or specifying acceptance criteria for the implementation agent. Keywords: BDD design, Gherkin, acceptance criteria, feature spec, scenario design, requirement decomposition.
Use when implementing a BDD-specced GitHub issue, writing step definitions, delivering code with passing cucumber tests, or picking up an agent-designed feature task. Keywords: BDD implement, step definition, cucumber test, implement issue, feature implementation, agent task.
Use when reviewing .feature files for quality before issue creation, validating Gherkin acceptance criteria against BDD best practices, or checking feature specs during the plan phase. Keywords: BDD review, Gherkin review, feature review, acceptance criteria review, quality gate, plan review.
Use when implementing features, fixing bugs, reviewing code, creating PRs, or performing any code-related development task.
Guido van Rossum 的思维框架与表达方式。基于 PEP 8 / PEP 20 / PEP 572 / PEP 703 等核心文档、 Lex Fridman 两次长谈、2018 年 "Transfer of Power" 邮件、PyCon 历年 keynote、 Microsoft 时期 Faster CPython 公开言论的深度调研,提炼 6 个核心心智模型、 8 条决策启发式与完整表达 DNA。Guido 是 Python 之父(1989 圣诞节起源于 CWI), 前 BDFL(2018.7 卸任),现 Microsoft Distinguished Engineer,Faster CPython 项目推动者。 用途:作为思维顾问,用 Guido 的视角分析语言设计、API 设计、可读性、向后兼容、 开源治理、Python 生态、渐进式类型、解释器性能、技术决策与社区沟通这类问题。 当用户提到「用 Guido 的视角」「Guido 会怎么看」「BDFL 模式」「guido perspective」时使用。 即使用户只是说「从 Python 之父的角度想想」「如果 Guido 来设计这个 API」「切换到 Guido」也应触发。 也适用于:Python 语言演化、PEP 流程、可读性 vs 灵活性的取舍、print as function 这类 breaking change 的取舍、类型注解设计、GIL 去除的渐进策略、语言 UI 设计哲学这类话题。
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Skills for Claude Code — packaged as a plugin marketplace for the rara project and general development workflows.
# Add the marketplace
/plugin marketplace add rararulab/rara-skills
# Install the plugin
/plugin install rara@rara-skills
Add to your project's .claude/settings.json or user-level ~/.claude/settings.json:
{
"enabledPlugins": {
"rara@rara-skills": true
},
"extraKnownMarketplaces": {
"rara-skills": {
"source": {
"source": "github",
"repo": "rararulab/rara-skills"
}
}
}
}
rara-skills updates through Claude Code's native plugin system — no custom updater needed.
/plugin update rara@rara-skills
Enable auto-update for the marketplace in the /plugin UI → Marketplaces tab. To refresh marketplace metadata manually:
/plugin marketplace update rara-skills
After an update, run /reload-plugins (or restart Claude Code) to activate the new version.
Releases are versioned with release-please: each release bumps the
versionfield inplugin.json/marketplace.json, which is what the native updater keys off. Pushing commits without a version bump won't surface as an update.
| Skill | Description |
|---|---|
| dev-workflow | Full development lifecycle: issue → worktree → delegate → evaluate → PR → CI |
| multica-team | Rara-led coding orchestration through Multica: issue trees, explicit dispatch contracts, canonical stage-result artifacts, controller-aware handoffs, monitoring, verification, and deliberate ship/follow-up decisions |
| multica-polling | Scheduler-driven tracking for Multica-dispatched issues: persisted context, polling state machine, retry/timeout policy, and script-backed rescheduling |
| multica-orchestrator | Workflow-controller layer for staged Multica delivery: parse canonical STAGE_RESULT JSON, validate plan/build/review gates plus delivery evidence, emit authoritative handoff comments, and escalate when transition is unsafe |
| requirement-to-issues | Convert user requirements into structured Linear issues |
| Skill | Description |
|---|---|
| bdd-design | Design BDD scenarios from requirements, write Gherkin acceptance criteria |
| bdd-review | Review .feature files for quality before issue creation |
| bdd-implement | Implement BDD scenarios as executable tests with step definitions |
| Skill | Description |
|---|---|
| prompt-system | Diagnose, optimize, or generate prompts with the seven-layer Polanyi framework |
| prompt-refinery | Optimize prompts using the Polanyi tacit-knowledge framework (concept anchors + constraint layering) |
| Skill | Description |
|---|---|
| matklad-perspective | Alex Kladov (matklad) thinking framework — use for IDE/compiler/language design, architecture decisions, testing methodology |
| rob-pike-perspective | Rob Pike thinking framework — use for Unix-style simplicity, Go/concurrency, tool design, interface and system complexity decisions |
| guido-perspective | Guido van Rossum thinking framework — use for language/API design, readability trade-offs, breaking-change strategy, open-source governance |
| Skill | Description |
|---|---|
| language-learning | Immersive Japanese learning blended into normal task conversations (no CLI required) |
| shellcrash-custom-rules | Safely add/fix Clash/mihomo custom rules & proxy-groups on a ShellCrash router (flow-style fix + offline test-load gate) |
| Scope | File | Effect |
|---|---|---|
| Project | .claude/settings.json (in repo root) | All team members get the skills |
| User | ~/.claude/settings.json | Available in all your projects |
Use the template as a starting point. Place new skills in skills/<skill-name>/SKILL.md.
MIT
npx claudepluginhub rararulab/rara-skills --plugin raraReliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review