Development plugins

Claude Code and Codex plugins their authors filed under "development". All categories.

disensor

claude

Run a cross-model adversarial review round with the disensor CLI and declare its residue.

2 skills

NicolasRocchia/disensor-claude 0

curlshot

claude

Screenshots of any public web page as PNG, JPEG, WebP, PDF or a scrolling video: the hosted CurlShot MCP server plus an agent skill for the HTTP API.

1 skill · 1 MCP server

SergiuPogor/curlshot-mcp 0

Skills for AI coding agents by anguspiv: angusp-voice, writing-issues.

2 skills

anguspiv/agent-skills 0

nwscript

claude

KotOR modding with nws and rakata: write, check, test and compile NWScript; find the copy of a resource the game loads, read and change dialogues, globals, tables and blueprints through the Python module, and check a change with the lints and a diff.

2 skills

ivy/nwscript 0

berth

claude

Berth rules for agents plus a guard hook that blocks device commands outside a berth lease.

1 skill · hooks

LunabeeStudio/berth 0

Writes code with you, and turns you into an engineer who can see at a glance where the AI got it wrong.

2 skills

jackeyloveseven/ai-coding-coach 0

quonfig

claude

Use Quonfig feature flags, dynamic config, and log levels from your application code: pick and wire up the SDK, create flags with the qfg CLI, and manage them through the Quonfig MCP server.

1 skill · 1 MCP server

quonfig/skills 0

Agentic verification for Python/pytest. Verify a change: run every mutant of the changed lines and write tests until the tests catch them. Audit a suite: find what it would not catch, what makes it too slow or flaky for an agent loop, and which tests can go without losing detection.

2 skills

timini/agentic-verification 0

Verosoft feature lifecycle for BC and Mobi: sizes the work, runs its steps, keeps one home for specs and stops for approval at named gates.

2 skills

Verosoft-Design/claude-plugins 0

Review each PR as an ensemble: challenge the problem, assumptions, and design, identify what code to delete while preserving all functionality, and check the implementation.

1 skill · 1 command

pierrebrunelle/challenge-the-design 0

Check GitHub Actions status using official incidents and unusual visitor interest, with clear evidence and stale-data timestamps. Public and read-only.

1 skill · 1 MCP server

openmerchants/openmerchant-plugins 0

Project instruction + slash command for building UASF v8 agents

1 command · 1 agent

jasonpiedrasantasdk/UASF-Universal_Agent_System_Format 0

no_human

claude

File a task with no_human and check on it without leaving Claude Code: task_add and task_status over your local no_human server (127.0.0.1:8420). no_human plans, codes, tests and has the work reviewed by a second model, then opens the pull request and stops.

1 skill · 1 MCP server

eyalgolan/nh-action-e2e-scratch 0

Automated full-stack software engineering. Routes any dev task through specialized domain agents (database, backend, frontend, testing, calls, bridge, git) in dependency order, asking plain-English questions first. Orchestrator runs on the host's main model; mechanical sub-agents run on cheaper models. No API keys or local Ollama required.

2 skills

MARafey/The-Software-Engineer 0

collaborate

claudecodex

Work with people in live revise.io documents over the Revise MCP server: tracked suggestions, comments, and shared editing.

1 skill · 1 MCP server

reviseio/skills 0

Reusable engineering workflows and delivery guardrails.

2 skills

natata7/dev-digest-ai-marketplace 0

sdlc

claude

Software delivery workflow: convention-aware commits and GitHub issues, PRs, reviews and PR listing through the gh CLI.

2 skills

provectus/claude-code-advanced-workshop-toolkits 0

backend

claude

Backend engineering: Kotlin skill and Kotlin engineer agent.

1 skill · 1 agent

provectus/claude-code-advanced-workshop-toolkits 0

review

claude

커밋 전에 Codex 리뷰를 받아 검증된 지적만 반영하고, 리뷰를 통과하기 전에는 커밋·푸시를 막는 게이트

1 skill · hooks

LimjeongSik/baton-workflow 0

feature

claude

구현 → 테스트 → 검증을 역할별 서브에이전트로 나눠 진행하는 기능 개발 워크플로우

2 agents

LimjeongSik/baton-workflow 0

Tobias Karlsson's implementation plugin: how names and comments read, and the pre-PR review cycle.

2 skills

mattiasthalen/claude-plugins 0

AI-native SysML v2 co-pilot: authors valid textual SysML v2 (.sysml), grounds every type/unit in the standard library (ISQ, SI, …), and validates with the real SysML v2 Pilot compiler in a self-correcting loop. Bundles an MCP server (validate_sysml_file, query_library, get_library_element, dump_model).

1 skill · 1 MCP server

3shn/skill-sysml 0

proofcheck

claudecodex

Verify completion claims with fresh, scoped evidence without changing the project.

1 skill · hooks

codewithasnari/proofcheck 0

engineer-project

claudecodex

Guide software projects from discovery through verified release and operations.

1 skill · hooks

codewithasnari/EngineeringPlaybook 0

按 codego 基线初始化 AI Coding 友好的 Go 项目(HTTP 服务或 CLI),并只读审计现有 Go 项目给出分级建议

2 skills

maguowei/plugins 0

shelf

claudecodex

The shelf skills — manage a personal shelf documentation library through the `shelf` CLI (add, find, edit, remove, audit, open, and set up documents), and research a resource link into an original HTML reference page.

2 skills

smorinlabs/shelf 0

Draft qSQL via the fine-tuned qwen specialist — /qsql-coder (with Bash-verify) and /qsql-coder-quick

2 skills

KxSystems/mcp-qsql-suggester 0

xreview

claudecodex

Have Codex review Claude Code's work and Claude Code review Codex's, read-only, from the real git diff; convene Claude, Codex and Gemini to deliberate on plans and results.

2 skills

Armin-Saadat/xreview 0

Start a new multiplayer browser game with a GitHub repo, CI, Render deploys and a CLAUDE.md workflow for parallel agent sessions: /new-game.

2 skills

steven008/game-starter 0

vibeco

claudecodex

Build and operate partner integrations for the Vibe Developer Platform.

2 skills

vibe-ad/public-skills 0