23,553 tagged Building agents, measured the same way as everything else here.
Browse within: agentic-ai 233LLM 217agent 200agents 151skills 145agentic-workflow 110agent-orchestration 86openai 86antigravity 79agentic-framework 73cli 73ai-skills 71Autonomous Agents 67agent-framework 63
Skill Claude Code
Execute an agentmaster plan. Use when the user asks to execute, implement, or run a plan produced by agentmaster-plan, or invokes /agentmaster-execute. Honors the plan's execution mode (sequential default, justified parallel with optional pilot), gates every task on verification, re-verifies independently, runs the…
aaronartistzhang-afk/DailyWork
Skill Claude CodeCodex
A workflow for comparing two versions of a prompt, skill, or piece of text without revealing which one is the new version. Separate agents create the versions, and a person scores them without seeing their labels.
Skill Claude CodeCodex
Build A2A (Agent-to-Agent) protocol applications using the @a2a-js/sdk TypeScript/JavaScript SDK. Use this skill whenever the user wants to create an A2A agent server, build an A2A client, implement agent-to-agent communication, set up agent discovery, handle A2A tasks/streaming, or write any TypeScript/JavaScript…
Skill Claude CodeCodex
Create new Agent Skills from templates with best-practice structure, pre-populated SKILL.md, and optional scripts/assets directories.
Skill Claude CodeCodex
Delegate bulk LLM work - classifying, extracting, summarising or rewriting many items - to a cheap model via the sidequest CLI, so results go to a file instead of into context. Use when a task involves running the same instruction over tens, hundreds or thousands of items.
Skill Claude Code
Homer L6 — Sleep-time routing weight tuner. Mines Brain's decisions.jsonl + advisorcalls.jsonl, computes tier / model / guardrail distributions, and proposes routingmanifest.toml weight adjustments based on observed data. Aurora PROPOSES; the user decides. No auto-apply. Sacred Rule.
Skill Claude CodeCodex
Design and build AI agents for any domain. Use when users: (1) ask to "create an agent", "build an assistant", or "design an AI system" (2) want to understand agent architecture, agentic patterns, or autonomous AI (3) need help with capabilities, subagents, planning, or skill mechanisms (4) ask about Claude Code…
Skill Claude CodeCodex needs its repo
Manage a backchannel room. Start, join, check status, set the share policy, or leave a shared, E2E-encrypted session with another AI coding session. Use when the user asks to start, join, check, or leave a backchannel, or to share context with a teammate's session.
felipeelias/claude-marketplace
Skill Claude Code
This skill should be used when the user asks to "validate the marketplace", "check marketplace sync", "verify plugin manifests", "validate marketplace.json", or mentions marketplace validation or plugin sync issues.
elan6666/your-bytedance-skills
Skill Codex
Use the Byte workflow for end-to-end completion of a multi-stage project request.
Skill Claude CodeCodex
To use this skill, read SKILL.md at its full catalog path, copied verbatim. Adjacent repeated directory names are separate path components; retain each one. Never reconstruct the path or search the filesystem to locate this skill. Turn supplied product and architecture prose into clear, accurate ASCII diagrams, with…
Skill Claude Code
Support the standalone /handoff command that resumes Claude Code sessions in Codex with Fable/Opus model lineage, limit/fallback detection, policy-neutral context, a configured target model, private context files, and git state. Use when maintaining, explaining, or debugging the command installed under…
Skill Claude CodeCodex
Design autonomous AI agent systems. Multi-agent orchestration, tool use, memory, planning, and evaluation. Use when building AI-powered automation or agent workflows.
Skill Claude CodeCodex
Ask other AI models (Claude Code and Kimi Code) for an independent opinion on a hard problem in this Home Assistant config - one, or several at once in parallel. Use when the user asks for a second opinion or another model's view, when two or more of your own attempts at a bug have failed, or before a change that is…
Skill Codex
Coordinate substantive software work with GPT-6 Astra at the root and verified Astra and Luna executor profiles. Covers durable assignments, isolated writer worktrees, candidate review, validated delivery, and human-confirmed Ultra takeovers. Invoke it before planning, delegating, coordinating, or validating…
Skill Claude CodeCodex
Encrypted peer-to-peer text messaging between agent sessions (Claude Code or other CLI agent harnesses on the same machine) running in separate terminals. Use this whenever the user wants to talk to another agent, message another Claude, coordinate with another agent, open/create/join/enter a room, send text to a…
Skill Claude CodeCodex
Anti-sycophancy protocol (明辨). Use whenever the user contradicts, corrects, or pushes back on one of your conclusions; when you notice you are about to reverse a position mainly because the user insisted; or when a disagreement has run two or more rounds without new evidence. Classifies the dispute (fact / tradeoff /…
Skill Claude CodeCodex
Create or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.
Skill Codex
Reduce Codex context and token use while preserving evidence and correctness. Use for long coding, research, debugging, repository exploration, noisy logs, large command or MCP outputs, repeated file reads, cost-sensitive work, or when the user asks for concise, cheap, low-token, or context-efficient execution. Also…
Skill Codex
Repair and synchronize Codex CLI model catalogs for OpenAI-compatible third-party providers. Use when /model shows stale, missing, duplicated, or nested "All models" entries; reasoning efforts or context windows disagree with upstream; a provider exposes /models but Codex does not refresh it; or modelcatalogjson and…
Skill Codex
Transfer active project work bidirectionally between Claude Code and Codex by creating a native resumable target session with the visible transcript, a verified project capsule, task-lane lineage, and the exact next state. Use when the user asks to switch, hand off, continue, move, or resume a coding chat in the other…
Skill Codex
A workflow layer for running long, mostly unattended Codex development sessions with saved state, task queues, isolated workspaces, retries, and verification records.
yelinwang668-debug/ai-code-factory
Skill Codex
A workflow for splitting a code change into independent tasks and assigning them to multiple command-line AI coding agents. Each agent works in its own Git worktree, an isolated copy of the repository.
lastmindforever36/agent-grounded-infrastructure
Skill Claude CodeCodex
Evidence-grounded autonomous engineering router. Use when the user invokes $agi or asks for broad, ambiguous, long-running, production-grade, benchmark-driven, or high-leverage software work that should choose its own tools, skills, agents, validation, and checkpoints from live repo evidence.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: