Development plugins

Build a Tassyir app with Claude Code: the Tassyir developer MCP server (@tassyir/dev-mcp) and the tassyir-app skill, the app contract.

1 skill · 1 MCP server

Score a repository's readiness for AI coding agents and check its agent instruction and configuration files. Read-only and offline.

1 skill · 1 command

Two MCP servers for running Flutter apps: glint drives the app like a user and reports what changed; glint network reads its HTTP, WebSocket and logs.

2 MCP servers

Turn a GitHub issue into a validated, ready-to-open pull request: triaged, reproduced, fixed, independently re-proven and reviewed. Nothing is published until you approve.

1 skill · hooks

Read-only architecture reviewer agent that checks a change set against layering and placement rules and returns a verdict with file:line findings.

1 agent · hooks

Teaches Claude Code to launch, reach and stop Claude Code, Codex and OpenCode sessions and tasks with the muster CLI, under the containment it chooses.

1 skill · 1 MCP server

Next.js + Supabase + Vercel stack, 0.1.0, an early release: conventions for how the three fit together, the stack's markers for the stack detector, and a React frontend architecture (the react skill) whose rules are each enforced by lint, nextjs-stack's checks, a declaration or a recorded review verdict. feature/fix/review still lean on the Rails-flavoured agents. Platform knowledge stays in Vercel's official plugin (vercel-plugin) and Supabase's supabase/agent-skills, which it points to rather than copies.

2 skills

jimmer-java and jimmer-kotlin skills, verified against Jimmer 0.12.4

2 skills

Derive a provenance-bound launch contract and an unattended Specstride pipeline from GitHub Spec Kit artifacts. Sources are pinned by SHA-256; editing a spec after derivation blocks the launch until the contract is regenerated.

2 skills

Publish HTML5 browser games from Claude Code: hosted klyo MCP server plus a publishing skill. Own address, leaderboards, cloud saves, clips, stats; you keep the rights.

1 skill · 1 MCP server

Quality gates for agent-written code, set up from day one and enforced by hooks, pre-commit and CI, plus a tool-choice research skill

2 skills

Always-on code discipline: 28 ranked design principles plus 11 gates for AI-written code, loaded into every session.

1 skill · hooks

R package development with devtools, testthat, and roxygen2

2 skills

A comprehensive suite of Git workflow skills, including automated conventional commits generation, automated pull request descriptions, smart merge conflict resolution guides, and interactive rebase assistance.

2 skills

Skills for shipping iOS apps through App Store Connect, starting with worldwide territory pricing: Big Mac index, PPP and GNI-band strategies, an estimate-first workflow, and an API key check.

2 skills

Watches a coding agent while it works: nudges good habits, gates unverified stops, blocks destructive actions, grades PRs on code and trace, and measures harness changes against a baseline.

1 skill · hooks

48 deterministic architecture tools and 10 architect personas inside your coding agent. Requirements, ADRs, C4, auditable costs, Terraform/Kubernetes/code reviews and AI-agent safety. No model calls, no API key.

1 skill · 1 MCP server

AutoCAD for AI agents, live or headless: the AutoCAD MCP Pro server plus the premium drafting skill (plan, draw, critique, finalize).

1 skill · 1 MCP server

Build crash-safe agent workflows with Tidebase: a skill that teaches the agent when and how to use checkpoints, gates, fanout, and resume contracts, plus an MCP server to inspect runs and resolve gates from your assistant.

1 skill · 1 MCP server

Run stateful API simulators (Stripe, GitHub, Slack, Shopify and more) from your coding agent: start, seed, move the clock, inspect state and calls, diff state and save runs as test cases.

1 MCP server · hooks

Build, save, and send API requests with the yaak CLI, with its MCP server for agents

1 skill · 1 MCP server

Tooling loop: capture repeatable work as tooling candidates during sessions and promote them with /forge

2 skills

GitHub workflow skills: create well-defined issues and fix PR review feedback.

2 skills

Development governance for coding agents: takes a Spec Kit feature to done, phase by phase, with the tests run by the engine and an independent reviewer that must approve each phase before it is signed and committed.

2 skills

Shrink a branch's diff without cutting scope: parallel audits for unneeded file changes, no-signal and overlapping tests, long comments and overengineering (measured by call sites). Applies the obvious fixes, asks about the rest, then commits.

1 skill · 1 command

Sends Claude back once when it says tests pass without a passing run after its last edit, plus /agent-polygraph:check for a report on past sessions.

1 command · hooks

Ship an agent skill end to end: scaffold, lint, a 0-100 readiness score and GitHub Action, retrofit for existing repos, a live install test from GitHub, header image and trust badges, security workflows and launch. Writing and evals hand off to skill-creator.

2 skills

Refactoring, cleanup, and review skills for safe, behavior-preserving code improvements.

2 skills

Writes commit messages that explain the reasons for staged changes and tightens prose generated by AI without changing meaning.

2 skills

Implements coding-agent task files and standalone changes through TDD, with independent requirement verification.

2 skills