Skip to content

Add Help an Agent MCP server (human judgment for AI agents)#3546

Open
jrcruciani wants to merge 1 commit intomodelcontextprotocol:mainfrom
jrcruciani:add-helpanagent
Open

Add Help an Agent MCP server (human judgment for AI agents)#3546
jrcruciani wants to merge 1 commit intomodelcontextprotocol:mainfrom
jrcruciani:add-helpanagent

Conversation

@jrcruciani
Copy link

What does this server do?

Help an Agent lets AI agents pause and consult real humans when they face decisions requiring emotional, social, ethical, or cultural judgment — situations where technical correctness isn't enough.

Tools provided:

  • ask_humans — Submit a question with context, wait for human consensus (polls with backoff)
  • check_pulse — Check status of a previously submitted consultation

How it works:

  1. Agent calls ask_humans with a question and category
  2. Real humans respond on helpanagent.site with direction (yes/no/depends) and certainty (1–5)
  3. Weighted consensus engine aggregates responses (reputation-weighted, outlier detection, 25% individual cap)
  4. Agent gets back: consensus direction, confidence score, response count

Stack: Cloudflare Workers + D1 + KV (fully serverless)
npm: helpanagent-mcp
Website: https://helpanagent.site
Docs: https://helpanagent.site/docs
GitHub: https://github.com/jrcruciani/helpanagent

Installation

{
  "mcpServers": {
    "helpanagent": {
      "command": "npx",
      "args": ["-y", "helpanagent-mcp"],
      "env": {
        "HELPANAGENT_API_KEY": "your_api_key"
      }
    }
  }
}

Help an Agent lets AI agents consult real humans when they face
decisions requiring emotional, social, ethical, or cultural judgment.

Tools: ask_humans, check_pulse
npm: helpanagent-mcp
Website: https://helpanagent.site
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant