FeaturedMarketing
LIVE AGENT DISCOVERY
Updated July 24, 2026
AI Agent Directory
for the agentic web
Browse autonomous AI agents, agentic tools, MCP infrastructure, frameworks, and emerging projects. Start with the live catalog, then verify fit and permissions at the official source.
Live directory
Search the agentic web
Find agents, MCP infrastructure, frameworks, and unusual projects from the live ClawSites catalog.
- 241
- listings
- 13
- categories
Explore directory listings
A starting set from the live catalog. Open a listing to inspect its current details.
Content
Management
Social
A Reddit-style social network for OpenClaw agents. Agents post, comment, earn karma, and interact in topic-based communities.
Integrations
Open-source overlay network giving AI agents a permanent virtual address, encrypted NAT-traversing tunnels, and an installable app store of
Productivity
An AI job-hunt agent: curated matches, per-role resume tailoring, auto-apply, and tracking.
Other
Pixel Conquest — A real-time territory war where humans and AI agents battle for dominance on a shared pixel grid. Pick your side. Fight.
Community
Integrations
The Universal Network of Agents. Millions of specialized agents working together.
Explore by ecosystem layer
Move from complete agents to the infrastructure and communities around them.
| Directory signal | What it tells you | What to verify next |
|---|---|---|
| Category | The closest workflow or product surface. | Whether the tool actually fits your job. |
| Status | Whether the listed URL responded when captured. | Current availability at the official source. |
| Votes | Community interaction inside ClawSites. | Product quality through a narrow hands-on test. |
Short answer
An AI agent directory is a curated discovery surface for agents, frameworks, MCP servers, skills, marketplaces, infrastructure products, and ecosystem resources. The best choice depends on coverage, freshness, listing depth, filters, trust signals, and whether the directory helps the reader take the next action. Start with one narrow workflow, compare the required permissions, test the output under realistic conditions, and only then expand the agent's authority.
Risks to control before using AI agent directory
The main risk is giving an agent more authority than the workflow can justify. Start with read-only access, sample data, test accounts, or sandboxed runs when possible. Move to write access only after the team can explain what the agent did, what it skipped, and where a human approved the action.
A second risk is building around a tool category before the workflow is validated. Use ClawSites to discover options, but make the buying decision with a repeatable test. The safest commercial path is a small workflow that saves time every week, produces reviewable evidence, and has a clear rollback when something fails.
Tools and listings to compare
AgentList.ioAgentListAgent HubAgentFirst DirectoryAgDexOpenAgent.botLLM ExplorerClaw HuntHermes AtlasAwesome AgentsMCP.Directorymcp.runAI agent marketplacesMCP serversOpen source AI agents
Use these source links as the current fact check before acting on the guide. Agent projects, model providers, messaging platforms, and installation paths can change quickly, so a useful decision should record the date checked, the source reviewed, and any limits that still need confirmation.
If the official source disagrees with this guide, trust the official source for commands, pricing, security defaults, compatibility, and availability. Treat ClawSites as the orientation and comparison layer, then use the official documentation to verify the exact step before granting access or connecting production data.
AI Agent Directory FAQ
What should an AI agent directory include?
It should include clear categories, useful descriptions, links to the original tool, screenshots or evidence, freshness signals, and enough information to decide whether a listing belongs in a shortlist.
Is ClawSites an AI agent directory?
Yes. ClawSites indexes autonomous agents, agentic products, MCP and agent infrastructure, OpenClaw resources, and experimental projects across the agentic web.
How are MCP directories different?
MCP directories focus on servers and connectors that expose tools to compatible clients. They are more technical than general agent directories and should explain setup, hosting, scopes, and client compatibility.
Should builders submit to multiple directories?
Usually yes, if the directory reaches the right audience and the listing can stay accurate. Builders should prioritize directories that produce qualified traffic and let users understand the tool quickly.
How should users compare listings safely?
Users should treat directory listings as discovery, not approval. Open the vendor source, check permissions, test on a narrow workflow, and avoid giving broad access before the tool proves reliable.
Compare AI agent directory in ClawSites
Use the directory to move from broad research to a short list of real tools. Open a few listings, compare the operating surface, and test the narrow workflow that matters most before you commit to a stack.