Development plugins

Experimental: budgeted test runs with a stop-after-green rule, a guard for pre-existing tests, shaped test output and project memory (skill, MCP server, hooks).

1 skill · 1 MCP server · hooks

Integrate the VisioOne indoor mapping SDK (@visioglobe/visioone) into a web project, migrate VisioWeb, VisioWeb Essential and VisioMove Essential V2 apps to VisioOne, and prepare bug reports for Visioglobe support.

3 skills

Playwright, Chrome DevTools and Context7 MCP servers.

3 MCP servers

Drive hive from an agent: sessions, splits, sending keys, reading output, waiting on programs. Set hive up, and move work over from tmux.

3 skills

Analyze legacy code (COBOL, CICS, JCL, BMS, DB2, VSAM, IMS and more) and write versioned, traceable BRD, FRD, SRD, RTM and gap reports, plus the React + Spring Boot + DB2 migration design.

3 skills

Run long-running software projects through paired Thinker and Coder agents, with every turn persisted to disk and past decisions retrieved from a local vector store.

1 skill · 2 agents

Scan your app for App Store rejection risks before submission, and resolve rejections with guided fixes.

3 skills

Guitar Alchemist DSL: music theory tools, chord analysis, tab search and GA Language commands. It runs GaMcpServer from the current directory, so it works only in a session started inside a ga checkout

1 skill · 1 command · 1 MCP server

Builds one work item (story, bug, task, or chore) from its ticket to tested commits, spec first. Copies the Behaviour spec of the ticket and locks one test file for each rule. Builds one scenario at a time with TDD. Proves the result with held-back tests, a mutation check, and fresh reviews. Sets the size from the change and the risk. Product questions go back to the product owner on the ticket.

2 skills · hooks

Google'ın resmî Gemini becerileri: gemini-api-dev, gemini-live-api-dev, gemini-omni-flash-api. Kaynak: google-gemini/gemini-skills.

3 skills

Timeback/QTI platform reference — prevents XML, rendering, and API integration bugs across AP course builds.

1 skill · 1 agent · hooks

Local, read-only structural code navigation for Claude Code

1 skill · 1 MCP server · hooks

Coordinate work projects, research, and competitions with presets, classic or unified workflows, configurable guards, operational hooks, and durable handoffs.

1 skill · 1 agent · hooks

Agent Skills for Apache Sling developers and users.

3 skills

The GenClass MCP server, a prompt hook that brings new findings to each prompt, and a skill for intermittent and timing-dependent frontend bugs.

1 skill · 1 MCP server · hooks

Read a Figma selection as a compact outline or CSS and check your build against it with a visual diff. Runs locally; nothing leaves your machine.

1 skill · 1 command · 1 MCP server

Integrate Prizerail prize draws into an app, check an integration end to end, and look up or verify draws.

3 skills

/scaffold-context for new bounded contexts, the adr skill, and PreToolUse guards: framework-free <bc>/domain, protected arch configs, no raw colors in .vue.

1 skill · 1 command · hooks

Set up and maintain a Loom Loyalty tenant with treadle, the API and the SDK.

3 skills

Lets Claude work in live Jupyter notebooks with a running kernel, not just the saved .ipynb file: it runs cells, sees plots, keeps kernel state and uses any of your envs as a kernel. Starts a local JupyterLab on demand, on a laptop or a cluster node. Not affiliated with Project Jupyter.

1 skill · 1 MCP server · hooks

Code quality skills: review, refactoring suggestions, and naming convention checks. Demonstrates multi-skill plugins and $ARGUMENTS usage.

3 skills

Team coding standards enforcer with a custom agent, supporting skills, and settings. Demonstrates agents + skills + settings working together.

2 skills · 1 agent

Author, evaluate, security-review, and autonomously improve Claude Agent Skills.

1 skill · 2 commands

Personal code review: deterministic CLI gathers diff/context/learnings; the active Claude session reasons over them.

1 skill · 1 command · hooks

Auto-commits and pushes every agent turn in opted-in repos: stage → secrets-scan → commit → push, with one-command undo, quiet batching, and PR mode. Strictly per-repo opt-in via /autogit on.

1 skill · 1 command · hooks

Install codegraph and graphify in a repository, route the agent to the right graph with near-zero hook overhead, and build the graphify graph on your own LLM backend (LiteLLM, Ollama, OpenAI-compatible) instead of agent subagents

2 skills · hooks

Run Claude Code subagents on other harnesses (Codex CLI, pi) so pstack arena runners and cross-judges can use non-Claude models.

1 skill · 1 agent · hooks

Cuts built-in WebSearch usage cost by bundling two remote-hosted web-search MCP servers with free plans (Tavily, You.com; server names prefixed wsplus-, API keys from MCP_* env vars) and injecting a SessionStart instruction that makes Claude use them first, one at a time until its rate limit or quota is hit, with the built-in WebSearch as the final fallback.

2 MCP servers · hooks

Polotno skills for SDK development, editable design production, and live documentation search.

2 skills · 1 MCP server

Agent skills for building integrations with ONLYOFFICE developer SDKs and tools: embed ONLYOFFICE Apps with the Embed SDK, build portal plugins with the Plugins SDK, and build React screens with the ONLYOFFICE UI kit.

3 skills