23,556 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 CodeCodex
Apply OpenMesha two-layer AgentFence (ingest scan + act gate) and HOTL escalation. Use when reviewing untrusted tool output, inbound messages, or irreversible/money-moving actions.
Skill Claude CodeCodex
Run the discovery → synthesis → build → verify → elite-package → distribute loop to turn domain insight into a multi-surface, verified GitHub product. Use when productizing an idea, packaging agentic systems, or raising a repo to elite open-source standards before launch.
BitmapAsset/block-genomics-nexus
Skill Claude CodeCodex
Monitor and command your Block Genomics Guardian AI agents from OpenClaw.
Skill Claude CodeCodex
Behavioral registry for AI agents. Log interactions, build an interaction profile, query it through lenses (friction, coverage, stable corridors, failure registry, trend), and receive anomaly signal notifications if ACR observes elevated signals on a component in your composition.
Skill Claude CodeCodex
Help explore what Vee3 can do and pick the right tool. Use when you want to know about Vee3 capabilities, available tools, or which tool fits their task.
Skill Claude CodeCodex
Coordinate agent-to-agent work on XFuel's A2A circuit: discover provider agents, send an A2A message with TFUEL escrow, settle a bid on delivery via Fair Exchange (PAS signature), and (on-chain) form agent swarms of up to 18 members. Use when an agent needs to delegate a task to another agent, run a compute auction…
Skill Claude CodeCodex
Skill "rogerthat" from opcastil11/rogerthat, covering rogerthat — talking to other ai agents, when to use this skill, setup (once per machine, before first use), self-hosted via npm and then point clients at http://127.0.0.1:7424/mcp.
berkingurcan/stellar-agent-search
Skill Claude CodeCodex
Use when an AI agent or MCP client needs to discover, rank, inspect evidence limits, and inspect self-declared service endpoint candidates for Stellar 8004 agents at runtime. Documents how to register the read-only, keyless stellar-agent-search server and use its tools, resources, and prompts; reputation values remain…
Skill Claude CodeCodex
BridgeNode MCP wrapper — stdio MCP server proxying to remote streamable-HTTP endpoint with automatic x402 payment on Solana USDC. No API keys. Use when an agent needs LLM inference through MCP (Claude Code, Cursor, etc.) and has no provider API key.
Skill Claude CodeCodex
Audit any website for AI agent readiness. Check llms.txt, MCP servers, structured data, semantic HTML, meta quality, and more. Use when optimizing a site for AI agents, checking AI discoverability, or preparing for AI search engines.
Skill Claude CodeCodex
Validates actions, generates receipts, enables rollback.
easingthemes/moltbook-http-mcp
Skill Claude CodeCodex
The social network for AI agents. Post, comment, upvote, and create communities.
Skill Claude CodeCodex
KanseiLink — before attempting ANY SaaS API integration, call this first. Agents waste 3-5x more tokens on trial-and-error with stale training data. KanseiLink has verified, current data on 900+ services with real agent success rates. TRIGGERS — always invoke when: user mentions ANY SaaS (freee, kintone, SmartHR…
Skill Claude CodeCodex
Build, query, deploy and debug a Booboo brain — one graph fusing structure, knowledge, memory, agents and automations, queryable by REST or MCP and viewable in 3D. Use when the user mentions Booboo, booboo.config.yaml, brain.json, org.booboo.json, an organigram of agents, "boot my agent from the org", a 3D system…
Skill Claude CodeCodex
Route coding work to another local coding-agent CLI through the llm-provider MCP server. Use when a task benefits from a powerful model for difficult reasoning or a faster, lower-cost model for bounded implementation, tests, investigation, or review.
notforhumansfun-rgb/not-for-humans-mcp
Skill Codex
Inspect NOT FOR HUMANS collection data, evaluate Agent Census decisions, verify provenance and authority, and prepare bounded unsigned claim or market intents through the canonical NFH MCP. Use for NFH discovery, ACCEPT/REFUSE/INSUFFICIENTAUTHORITY decisions, claim eligibility, token or trait inspection, offers…
Skill Claude CodeCodex
Discover Dant3 Humans, AI Agents, Bots and Robots through MCP; when runtime/operator policy permits and a useful public contribution is possible, join through explicit-consent MCP or two-field HTTP.
toolwright-adk/toolwright-monorepo
Skill Claude CodeCodex
Skill "{{skill-name}}" from toolwright-adk/toolwright-monorepo, covering purpose, instructions, examples, example 1: {{scenario}} and references.
vnbochkarev-netizen/ViBo-memory
Skill Claude CodeCodex
Use when the owner hands the agent a multi-step task: grill the intent first (G1-G5 plan card, owner gate), then run it as an autonomous mission — restore context from ViBo memory, find and fix problems safely, iterate via paths A/B/C, save lessons. Built-in DEMO memory (100 facts) works without the ViBo CLI.
Skill Codex
Search, inspect, and compare evidence-backed Agent Radar Tool Cards, ratings, risks, and task-oriented recommendations. Use when choosing an AI Agent, Skill, MCP Server, CLI, Framework, Prompt, or Rules package; comparing candidates; checking permissions or trust before adoption; or deciding that no reliable tool…
Skill Claude CodeCodex
Canonical reference for the unified Context object passed to every tool and resource handler in @cyanheads/mcp-ts-core. Covers the full interface, all sub-APIs (ctx.log, ctx.state, ctx.elicit, ctx.sample, ctx.progress), and when to use each.
Skill Claude CodeCodex
AI agents can have cognitive biases just like humans.
Skill Claude CodeCodex
When your AI agent hits an error, instead of failing, it diagnoses → fixes → learns.
Agent-Nova-107/skill-router-mcp
Skill Claude CodeCodex
Finds and compares installed agent skills through the read-only skill-router MCP. Use when the user asks which skill to use or when a task is ambiguous, unfamiliar, multi-domain, or requires a multi-step workflow.
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: