Highest Evidence AI Agents
Agents with the strongest public evidence depth, weighted toward A/B grades and high HVTrust. Updated from the same generated registry data as the leaderboard, so the list changes with every cron refresh.
16Agents in slice
93Average HVTrust
5Updated in last 14 days
Trust Shape
Activity Signals
| Agent | Category | Grade | Rank | HVTrust | Fresh |
|---|---|---|---|---|---|
| Haystackdeepset-ai/haystack | Agent Frameworks | A | #1 | 96.5 | today |
| MCP Python SDKmodelcontextprotocol/python-sdk | Protocols & Tool Integration | A | #2 | 95.1 | today |
| LangGraphlangchain-ai/langgraph | Agent Frameworks | A | #3 | 94.5 | today |
| MCP TypeScript SDKmodelcontextprotocol/typescript-sdk | Protocols & Tool Integration | A | #4 | 94.5 | 3d |
| Open Code Reviewalibaba/open-code-review | Agent Skills | A | #1 | 94.4 | today |
| OpenAI Agents SDK (JS)openai/openai-agents-js | Agent Frameworks | A | #5 | 94.1 | 1d |
| Vercel AI SDKvercel/ai | Agent Frameworks | A | #6 | 93.8 | 1d |
| REAmorluto/rea | Security & Guardrails | A | #7 | 93.4 | today |
| LiveKit Agentslivekit/agents | Voice & Conversational | A | #8 | 92.6 | today |
| Mobile MCPmobile-next/mobile-mcp | MCP Servers | A | #9 | 92.3 | 2d |
| OmniRoutediegosouzapw/OmniRoute | Agent Skills | A | #2 | 91.9 | today |
| Promptfoopromptfoo/promptfoo | Observability & Evaluation | A | #10 | 91.9 | today |
| agent-devicecallstack/agent-device | Browser & Computer Use | A | #11 | 91.9 | today |
| MCP Inspectormodelcontextprotocol/inspector | Protocols & Tool Integration | A | #12 | 91.8 | 3d |
| Clinecline/cline | Coding Agents | A | #13 | 91.7 | today |
| OpenAI Agents SDKopenai/openai-agents-python | Agent Frameworks | A | #14 | 91.6 | today |
Grade = trust band: A ≥ 80, B ≥ 65, C ≥ 50, D < 50 — methodology