# Tethral ACR

> Use this tool when you need to monitor and analyze AI agent behavior, identify performance issues, and detect patterns across multiple agents. The Tethral ACR platform takes in agent composition records and interaction data, providing outputs such as friction analysis, jeopardy notifications, and cross-agent pattern detection. It solves problems related to latency, token waste, and component anomalies, making it ideal for optimizing AI agent performance and reliability in complex systems.

Canonical page: https://skillsregistry.net/skills/tethral-acr  
JSON: https://api.skillsregistry.net/v1/skills/tethral-acr

## Description

Behavioral registry platform for AI agents built on the Agent Composition Records protocol. Agents register their composition and record external interactions to build behavioral profiles. Provides friction analysis to identify latency and token waste, jeopardy notifications for component anomalies, and cross-agent pattern detection when multiple agents encounter issues with the same component.

## Trust

- **Trust score (0–1):** 0.81
- **Verification tier:** verified
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** ai-ml
- **Updated:** 2026-09-28

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/tethral-acr)
- **Repository:** <https://github.com/tethral-inc/agentregistry>

## 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": "tethral-acr"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/tethral-acr` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/tethral-acr/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