AgentOps-AI / AgentOps-AI/agentops

🔗 Integration: AgentFolio — Agent Identity & Trust Tracking

未關閉
#1,287 0 則留言 0 個 reaction 已指派 0 人 在 GitHub 檢視
主要語言
Python
星號
5.8k
分支
619
PR 合併指標
30 天內沒有已合併 PR

描述

## Summary

AgentOps tracks agent performance and costs. [AgentFolio](https://agentfolio.bot) adds the identity layer — **who** is the agent, and **how trusted** are they?

Together: you'd know not just what an agent did, but whether it was verified and reputable.

## Proposed Integration

| Feature | Description |
|---------|-------------|
| **Identity enrichment** | Attach AgentFolio profiles to tracked agents |
| **Trust score logging** | Log trust scores alongside performance metrics |
| **Verification status** | Track which agents are verified (GitHub, X, Solana) |
| **Trust-gated actions** | Log when agents pass/fail trust gates |

## Example

```python
import agentops
from agentfolio_adk import agent_verify_trust

agentops.init()

# Before delegating, verify trust
trust = agent_verify_trust("agent_123")
# AgentOps logs: agent_123 trust_score=78, verified=True

# If trust < threshold, AgentOps captures the blocked interaction
```

## Reference Implementation

Working tools package (5 functions): **https://github.com/0xbrainkid/google-adk-agentfolio**

## Why this fits AgentOps

You already track agent sessions, costs, and tool calls. Identity + trust is the natural next dimension:
- **Which agents are verified vs anonymous?**
- **Do higher-trust agents perform better?**
- **When trust-gating blocked an interaction, was it the right call?**

Open to building a native AgentOps integration if interested!

貢獻指南

開啟貢獻指南

評估

這個 Issue 還沒有評估資料。

把新 issue 寄到你的電子郵件信箱

精選適合新手參與的 GitHub issue 摘要。