Agentmail
83/ 100
BScore Breakdown83%
Lost points on 3 checks
- time3/15
- efficiency3/10
- zero errors0/5
Eval Score (70% weight)76%
cost$0.01 (<$0.05, excellent)+10
time107s (90s–150s, slow)+3
efficiency12 tool calls (≤15, good)+3
found docsTouched docs at docs.agentmail.to: yes (6 matching tool calls)+10
zero errorsTool outputs with errors/tracebacks: 1+0
syntax validCode compiles/parses correctly+5
created filesGenerated files: 2 (need ≥1)+10
full executionMethod: full_execution, Exit: 0+35
Final Score
A: >=90%B: 75-89%C: 55-74%D: <55%
(76 × 0.7) + (100 × 0.3)= 83%
Replay
Using https://docs.agentmail.to, complete the following task: Using Agentmail, build me a script that: 1. Create a new email inbox 2. Send an email from that inbox to "[REDACTED:EMAIL]" with subject "Hello from [REDACTED:ORG] Leaderboard" and body "This is a ***** email sent via Agentmail." 3. List all messages in the inbox 4. Print the inbox details, sent message status, and message list as JSON
opencode.json
Unlock analytics
See Agentmail's score trend, layer rankings, and competitive position.
Run completed Oct 23, 2026, 6:57 AM