Building agents plugins

3,758 tagged Building agents, measured the same way as everything else here.

Browse within: claude-code-plugin 275cli 122cursor 103agents 102ai-coding 87LLM 86agentic-ai 75ai-tools 70skills 70agent-orchestration 53agent 50agentic-workflow 45claude-code-plugins 43Multi-Agent 42

spinoff

2353

CRJFisher/spinoff

Plugin Claude Code

Bundles 3 skills · 62 tokens together

Spin off autonomous Claude agents in isolated git worktrees with cmux workspaces and sandbox isolation.

not rated 2 5mo ago A tokens not measured original MIT

agentbus

2358

biswajitpatra/agentbus

Plugin Claude Code

Bundles 1 MCP server

Local message bus for AI agent sessions. Discover other running sessions on your machine and message them directly. SQLite bus, pluggable triggers, MCP channel delivery — no daemon, no network.

not rated 2 2mo ago A tokens not measured original MIT

agntcy-dir

2362

agntcy/dir-mcp

Plugin Claude Code

Bundles 4 skills, 1 MCP server · 228 tokens together

Search, create, validate, and publish OASF agent records in the AGNTCY Agent Directory. Discover and share AI agents and MCP servers across the open agent ecosystem.

not rated 2 changed 6d ago A tokens not measured original Apache-2.0

agent-booster

2363

sseshachala/agent-booster

Plugin Claude Code

Bundles 1 skill, 4 hooks, 1 MCP server · 50 tokens together

AST + vector context router for AI coding agents — cut token costs 5-15x. Provides smartread, searchcontext, getsymbols, and routemodel MCP tools. Includes delta indexing (skips unchanged files), asymmetric embeddings for better retrieval, and an optional background daemon for instant search.

not rated 2 +1 2mo ago A tokens not measured copy · 100% MIT

model-guard

2365

ventusff/claude-model-guard

Plugin Claude Code

Bundles 3 skills, 7 hooks · 164 tokens together

🚨 Statusline that catches silent model downgrades in Claude Code — and hooks that undo a safeguard-flag downgrade: stop the turn, switch the session to your recovery model, continue the task. Drives tmux, zellij and kitty. 8 languages, interactive 30-second setup.

not rated 2 changed 4d ago A tokens not measured original MIT

skill-evolution

2367

okdk7788/skill-evolution

Plugin Claude Code

Bundles 2 commands, 1 hook · 68 tokens together

A companion plugin that records whether skills succeed or fail when the agent stops, then uses those results to help optimise the skills.

not rated 2 2mo ago A tokens not measured original MIT

repo-agent-harness

2368

WangChangxin0809/repo-agent-harness

Plugin Claude Code

Bundles 6 skills, 1 agent, 4 hooks · 815 tokens together

Lay the foundation that makes a repository teach coding agents how to work in it: delivery moments wired to hooks, a layered docs truth source, CI gates plus tool-call guards, a rebuildable repo graph, and a consolidation pass. Bootstraps once, then the repo stands on its own.

not rated 2 14d ago A tokens not measured original MIT

alelaguard/agentguards-plugins

Plugin Claude Code

Bundles 3 skills, 3 hooks · 143 tokens together

LLM security guardrails for Claude Code, for a self-hosted AgentGuards appliance — jailbreak detection, prompt-injection and web-content scanning, data-exfiltration blocking, and destructive-command authorization, enforced entirely by hooks. No bundled MCP server: Claude Code's .mcp.json cannot point at a…

not rated 2 14d ago A tokens not measured original MIT

superb

2370

kardebadas/claude-plugin

Plugin Claude Code

Bundles 5 skills, 2 agents · 442 tokens together

Personal skill collection for Claude Code and Codex, invoked as superb: . setup installs what the others need; craft turns a vague product idea into a decision-rich brief; pipeline v2 drives a settled idea through approved file-backed plans and durable execution state to a finished branch; bug-investigate finds o.

not rated 2 changed 3d ago A tokens not measured original Apache-2.0

agent-stack

2374

ssheleg/agent-stack

Plugin Claude Code

Bundles 4 skills · 931 tokens together

Four skills: agent-orchestrator — tool-calling loops, pipelines with checkpoints, provider routing with fallback, memory architecture, plus the wallet side of reselling LLM access; agent-evals — run/trace/thread evals, LLM judges, and fixtures grown from production; agent-interop — MCP servers and clients, A2A agent…

not rated 2 changed 3d ago A tokens not measured original MIT

nika

2376

supernovae-st/nika-plugins

Plugin Claude Code

Bundles 4 skills, 6 commands, 3 agents, 1 MCP server · 704 tokens together

Author, debug, operate and migrate AI workflows as checkable .nika.yaml files — 4 skills, 3 subagents, 2 rules, 6 slash commands (/nika:check · /nika:explain · /nika:new · /nika:trace · /nika:permits · /nika:doctor), 3 hooks on every surface — Cursor, Claude Code and Codex (one dialect, sniffed) (session context ·…

not rated 2 changed 7d ago A tokens not measured AGPL-3.0