TriedAgents
Look up AI agent use cases that were re-run, scored out of 10 and graded by access level
- Pricing
- Free
- Platforms
- Web, API
- Runs
- Remotely
- Version
- 1.0.0
- Listing updated
- 1 day ago
About TriedAgents
TriedAgents is a free MCP server for ai models and llm tooling. It runs as a hosted remote server, so there is nothing to install locally.
TriedAgents takes claims about what AI agents can do, mostly from posts on X, and re-runs each with the exact prompt. Results are checked against ground truth and published with a verdict, a score out of 10 and an access level from L0 (public web only) to L4. Failures are included.
At the time of the page, the site reports 67 cases run: 30 reproduced, 32 partly reproduced and 5 failed, with two awaiting testing. Entries are specific, such as whether a founder built a site with a coding agent in about two hours or whether a prescription drug seller's prices match a pharmacy chain, and each shows the date and how long the agent took.
The MCP server makes those verdicts, scores, access levels and links queryable from ChatGPT, Claude or Cursor, so an assistant can check whether an agent claim holds up before you grant it access to accounts. A daily email with one re-run case is optional.
Install
Run in a terminal
claude mcp add --transport http triedagents https://triedagents.com/mcp/Add to Settings → Connectors → Add custom connector
https://triedagents.com/mcp/Add to ~/.cursor/mcp.json
{
"mcpServers": {
"triedagents": {
"url": "https://triedagents.com/mcp/"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"triedagents": {
"type": "http",
"url": "https://triedagents.com/mcp/"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"triedagents": {
"serverUrl": "https://triedagents.com/mcp/"
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"triedagents": {
"type": "http",
"url": "https://triedagents.com/mcp/"
}
}
}What it does
- Verdicts
- Each case is Reproduced, Partly reproduced or Failed.
- Scores out of 10
- Rates how well the claim held up.
- Access levels
- Grades the access an agent needed from L0 to L4.
- Failures published
- Cases that did not reproduce are listed too.
- Source links
- Entries link back to the evidence.
Good for
- Check an AI agent claim before relying on it
- Find cases that needed only public web access
- Compare how often agent claims fully reproduced
- Pull a verified example for an article
Pricing
Free
Free
Prices change. Check triedagents.com before you buy.
TriedAgents FAQ
What do access levels mean?
They run from L0 (public web only) to L4; the site has a page explaining scoring.
Are failures shown?
Yes, the site states failures are published.
Where do cases come from?
Mostly claims from posts on X.
Is it free?
The site and the daily email are free.
Is TriedAgents free?
Yes. TriedAgents is free to use.
How do I add TriedAgents to Claude Code?
Run this command in your terminal: claude mcp add --transport http triedagents https://triedagents.com/mcp/
What are the best alternatives to TriedAgents?
The closest alternatives to TriedAgents are Claude Code Ultimate Guide, signafAIde, OpenNash Podcasts, Toolkien and Lugh.
Alternatives to TriedAgents

Claude Code Ultimate Guide
MCP serverMCP server for searching the Claude Code Ultimate Guide from any MCP client
Runs locally- Open source
- 6.5K GitHub stars
- AI models and LLM tooling

signafAIde
MCP serverLocal MCP server to set up, verify and check coverage of signed agent action records
Runs locally- 5 tools
- Freemium
- Security

OpenNash Podcasts
MCP serverSearch business and tech podcast transcripts and sourced quotes over MCP, no key needed
- 10 tools
- Remote
- Free
- News and media

Toolkien
MCP serverOne MCP gateway for all your MCP servers, with code mode, permissions and audit
Not checked yet- 3 tools
- Remote
- Open source
- AI models and LLM tooling
Lugh
MCP serverWrite, check and dry-run Lugh agent programs with policy, budgets and approvals
Runs locally- 5 tools
- Open source
- AI models and LLM tooling
HypePaper
MCP serverResearch papers, benchmarks, citations and topic experts for AI assistants
- 58 tools
- Remote
- Freemium
- AI models and LLM tooling