# HumanProof

> HumanProof — sandeep-alluru-humanproof. Use this tool when you need to evaluate the human-likeness of automated workflows and identify bot-like patterns. HumanProof scores agent trajectories to detect anomalies, solving problems related to workflow automation and bot detection. It takes in agent trajectories as input and outputs a score indicating the likelihood of human-like behavior, making it ideal for use in production environments where automated workflows require validation.

Canonical page: https://skillsregistry.net/skills/sandeep-alluru-humanproof  
JSON: https://api.skillsregistry.net/v1/skills/sandeep-alluru-humanproof

## Description

Agent trajectory scorer for human-likeness — flags bot-like patterns in automated workflows before they reach production.

## Trust

- **Trust score (0–1):** 0.68
- **Verification tier:** scanned
- **Last scanned:** 2026-09-03

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** other
- **Updated:** 2026-09-03

## Source

- **Source listing:** [Glama](https://glama.ai/mcp/servers/bbc7zvos81)
- **Repository:** <https://github.com/sandeep-alluru/humanproof>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "sandeep-alluru-humanproof"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/sandeep-alluru-humanproof` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/sandeep-alluru-humanproof/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt