Development plugins

Build agents for any use: agent architect, orchestrator, prompt engineer, evaluator, MCP tool builder, plus the lean-agent-authoring skill.

1 skill · 6 agents

Companion of the Cross-Stack Standard: the Chrome DevTools MCP server with the privacy flags (isolated profile, no usage statistics, no CrUX lookups, redacted network headers) and its seven skills. Pinned to v1.10.1.

7 skills

Best practices, checkout, subscriptions, webhooks, Stripe migration and go-live review for Frisbii Billing and Frisbii Pay.

6 skills · 1 MCP server

コード・設計・テスト・レビュー・文章の品質基準を提供するスキル集。

7 skills

Testing vocabulary and standards. A guide that defines the terms for talking about tests and how to choose which kind to write, the RSpec base standards every spec follows with the unit, integration, and system standards built on them, and test debugging and TDD refactor step skills, each citing the guide for the reasoning.

7 skills

MCP tools and skills to install Mockingbird, configure an LLM, and create, serve and self-host mock APIs.

6 skills · 1 MCP server

Spec → Context → Build → Test → Evaluate → Deploy workflow for agentic coding, for new and existing repos

5 skills · 1 agent · hooks

Complete Knowledge, Memory, Experience, and Codebase tools in one Craft entry point

5 skills · 1 MCP server · hooks

Accessibility (WCAG) and UX checker for Claude Code: catches missing labels, contrast and keyboard problems after every edit.

6 skills · hooks

Claude Code leads, Codex and Antigravity work: briefs in, compact JSON results out, live progress in the agent panel, routing by task difficulty and each subscription's remaining quota

3 skills · 3 agents · hooks

Svelte 5 and SvelteKit 3 development, with a guided migration for SvelteKit 2 and Svelte 4 projects: current best practices, docs lookup and autofixer, language-server navigation with renames proven by the project check, an editor and an auditor agent, and the Svelte MCP and language servers, built on the Svelte team's AI tools.

3 skills · 2 agents · 1 MCP server · LSP

From issue to main: plan a change as an issue, commit it in small Conventional Commits and stacked pull requests, and land the stack by fast-forward.

7 skills

Hooks that run the repository's own lint gates on each file Claude edits and on the branch when it stops, a skill that splits a file over the line cap, and one that keeps vague names and fillers out of new code and prose.

7 skills

Workflow skills and guards: orienting in an unfamiliar repo, recording decisions with the evidence behind them, turning a feature or a finished brainstorm into spec rules and a milestone, building a milestone of spec rules task by task in the main session, fixing bugs test-first from docs/issues/, reviewing changes against their specs with an isolated reviewer agent, and hooks that hold the rules every time.

5 skills · 1 agent · hooks

The thinking before the code: grill a plan, keep the domain model in GLOSSARY.md and ADRs, design deep modules, surface deepening opportunities, prototype logic and layouts, research primary sources

7 skills

Skills de hardware, assembly 65816, ROM hacking, desmontagem, gráficos, testes e técnicas da época do Super Nintendo.

7 skills

Find, verify, install and run QuickAgents (signed, scanned agent packages with a declared capability contract) and build your own, from your agent host.

1 skill · 5 commands · 1 MCP server

Build and operate partner integrations for the Vibe Developer Platform.

7 skills

Skills for authoring, testing and auditing Inspect AI evaluations.

7 skills

A committed, searchable code graph for each repo, so an agent finds code without mapping the codebase again every session. Links paired repos to track feature parity.

6 skills · hooks

Report every open pull request a developer must act on in a GitHub organization — review requested, authored or assigned — as a raw table with clickable links, merge status, the concrete reason each PR is stuck, and age.

7 skills

Rank the unresolved Sentry issues of a project by impact — deterministic severity score, trend, innermost in-app frame, failure origin, root-cause clusters — then decide what to fix first by severity and complexity.

7 skills

Watch one pushed branch until its pull request is green, merged or blocked: rebase on the default branch, fix what CI and the reviewers report, one guardian per branch with a write lock against the human, orphan re-adoption at session start, and --dry-run for the plan without any push.

7 skills

Skills that teach coding agents to integrate QRGen scanning into any app: camera scanners, batch scanning, ID scanning, barcode generation, GS1/AAMVA parsing, migrations from other scanner libraries, and camera-free testing.

7 skills

Interviews the author, scaffolds a skill in a remembered local collection checkout, and registers its discovery metadata.

7 skills

Index stacked pull requests across an org and write the message that asks for review.

7 skills

Simplifies source code comments and docstrings into domain invariants and design intent. Excludes external Markdown documents and RFCs.

7 skills

Hand Rush a task: an agent does the work in its own cloud computer and returns a pull request, a report, files or an answer, with a record of what it did.

6 skills · 1 MCP server

Developer toolkit (sem-powered): annotate code with durable SEM@<sha> intent markers (sem-annotate), keep them maintained via a project convention (sem-auto), and find dead code + duplication and produce a ranked plan (dedupe). Uses the sem CLI entity graph. Supports Go, TypeScript/JavaScript, and Python.

3 skills · 3 agents

Build, test, deploy and debug AI agents on Fruxon: draft agents, connect tools, run evaluations, publish revisions and read execution traces

6 skills