Building agents

luminoguru-pvt/phasecraft

Skill Claude Code

Run Phasecraft SDLC. Auto-init if needed. Verifier triages fast vs interview; Planner compiles TASKCONTEXT + skills; Coder loads only selected skills.

kksudo/AgentFS

Skill Claude CodeCodex

Push the LLM to reconsider, refine, and improve its recent output. Use when user asks for deeper critique or mentions a known deeper critique method, e.g. socratic, first principles, pre-mortem, red team.

597844571/coding-skills

Skill Codex

Use when Codex should coordinate software project work through a virtual high-standard product and engineering team, including PM, product manager, UI designer, art director, architect, frontend engineer, backend engineer, algorithm engineer, QA engineer, and DevOps engineer. Applies to new projects, existing codebase…

girdav01/SoD-AISLDC-Skills

Cursor rule Cursor

SoD enforcement for Cursor 3 Agent mode, Background Agents, and multi-agent workflows. Activated when using agent features, background tasks, or parallel agent execution.

Xclaw-bot/benchmark-task-authoring

Skill Claude Code

Design, red-team, ship and debug hard Terminal-Bench 2 / Harbor benchmark tasks: the measured laws for what makes agents actually fail, the kill-list of dead task shapes, and how to clear all 17 review stages in one push instead of three. Use for benchmark task slots, TB2/Harbor tasks, task.toml, instruction.md, task…

stlwolf/ai-development-hub

Skill Claude CodeCodex

A lightweight guide for starting and coordinating a document-processing workflow. It provides fixed sections for delegation briefs, cold-start steps for new repositories or sessions, and a map of workflow stages and checkpoints.

setavenger/agent-skill-set

Cursor rule Cursor

Delegate context-heavy work to subagents; keep main session lean.

kingoo123/agent-harness-skills

Cursor rule Cursor

Agent Harness Skills. Use when designing AI agents, tool contracts, context pipelines, permissions, verification loops, or multi-agent systems.

activetwist/amphion-agent-extension

Plugin Claude Code

Bundles 9 skills, 1 hook · 277 tokens together

Scaffold operator-gated Micro-Contract Development workspaces with governance docs and a local Command Deck.

Halo-bo/work-review

Skill Codex

Review recent vibe coding or AI-assisted work history across Codex, Claude Code, WorkBuddy, Cursor, GitHub, local repositories, notes, or other available tools to identify repeated manual workflows worth packaging into reusable prompts, skills, slash commands, custom agents, playbooks, or automations, and infer the…

AIHongDOU/DoctorWhy

Cursor rule Cursor

DoctorWhy, medical requirement prober. When a medical software requirement is vague, ask it clear with one small question at a time until both sides understand, then produce runnable code.

GTMpath/Vibe-Coding

Cursor rule Cursor

AI behavior rules for generative features.

lizTheDeveloper/ai-agent-messaging

MCP server Claude CodeCodexCursor +2

MCP server "agent-messaging" as configured in lizTheDeveloper/ai-agent-messaging. Runs mcp_server/servers/agent_messaging_server.py with python. Needs 2 environment variables to run.

harness-mini/harness-mini

Cursor rule Cursor

This project runs on harness-mini. For non-trivial work — a feature, a multi-step or ambiguous change, anything cross-cutting — route it through the stage-viewer skill first and prefer harness-mini's lifecycle, skills, and sub-agents (read .claude/skills/ /SKILL.md) over ad-hoc tools or other installed plugins. When a…

diegoluchetti/magnifica_humanitas

Skill Claude CodeCodex

Use when starting any user interaction, especially when AI, automation, data, product, governance, persuasion, labor, education, security, or political requests may affect human dignity, truth, freedom, work, vulnerable people, power, or the common good.

mcku/ai-sdlc-orchestrator

Agent Claude Code

Subagent that runs the AI-SDLC framework end-to-end in an isolated context. Use when delegating a full SDLC session to keep the main conversation lean. The subagent boots from .ai-sdlc/ENTRY.md and executes phases sequentially, returning a summary on completion.

BilLogic/design-harnessing-kit

Cursor rule Cursor

SDS Compound Designing agent orchestration.

exponen-agi/recursive-agentic-improvements

Command Claude Code

Build any AI agent for any domain in any supported framework. The skill researches what tools and patterns actually exist before writing a single line of code.

hdk10/orgkit

Plugin Claude Code

Bundles 10 commands, 3 hooks · 420 tokens together

Turn a Claude Code repo into a self-organizing org: per-role memory brains auto-injected at session start, inline lessons auto-promoted, and a live ORG.md. Ships slash commands + lifecycle hooks plus a one-command onboarding flow.

HariEshwar-J-A/info-sentry

Agent Claude Code

Supergroup interaction agent for Info-Sentry. Listens in the Feedback forum topic, processes inline button callbacks from Main-News and Predictions, and lets the user manage their tracked interests through natural language. Runs on DeepSeek V3.2.

visionTw/godot-ai-harness

Command Claude Code needs its repo

A command that reviews recent Git commits and creates a brief of possible lessons to move into a shared AI harness, a reusable set of development instructions and tools.

jonZlotnik/how-to-agent

Plugin Claude Code

Bundles 27 skills · 1,477 tokens together

A library of Agent Skills (SKILL.md open standard) — composable lessons on decision-making, debugging, hygiene, defensive design, Unix-style composition, and technical writing. Portable across any skills-compatible agent. Adapted from Ada Jane Anderson's pragmatic-programmer-skills.

rsaglobaltech/spec-driven-development-template

Plugin Claude Code

Bundles 1 hook, 1 MCP server

Specs as executable contracts: the daily loop as commands, the spec tree over MCP, and the gate as a hook that runs before the session can end.

SrulyRosenblat/skill-cards

Skill Claude CodeCodex

Comprehensive guide for building AI agents using ToolLoopAgent, workflow patterns, and AI SDK UI components (useChat, generative UIs, tool calling).