See what your AI agents
actually do
Drop-in observability for Claude Agent SDK. One line change. Full traces, tool calls, and costs.
# Before
from claude_agent_sdk import query
# After (that's it)
from agentd import queryEverything you need to observe
Drop-in observability with zero configuration. Start simple, scale when ready.
Full Traces
Every tool call, input, output, and duration. Know exactly what happened.
Cost Tracking
Real-time cost per run, reported directly from Claude SDK.
Open Spec
Self-host with our reference receiver, or use our platform. No lock-in.
Zero Config
Works out of the box. Console output locally, HTTP when you're ready.
Rich console output
See everything your agent does, right in your terminal.
============================================================
📊 AGENT RUN SUMMARY
Run ID: 550e8400-e29b-41d4-a716-446655440000
Agent: my-research-agent
Status: completed
Duration: 12,340ms
Tokens: 15,230 in / 3,200 out
Cost: $0.0936
📦 Tool Usage:
• WebSearch: 3x (1200ms) [✓]
• Read: 2x (50ms) [✓]
• Write: 1x (30ms) [✓]
Get started in 30 seconds
Three simple steps. No configuration required.
Install
pip install agentd-sdk
Change one import
from agentd import query
Run your agent
See everything
Built on principles
We believe in transparency, simplicity, and developer freedom.
Open spec
The API contract is public. Build your own receiver or use ours.
No lock-in
Remove AgentD and your agent still works. It's just a wrapper.
Start simple
Console output by default. Add infrastructure as you need it.
The SDK is just the start
AgentD Platform is coming — hosted observability and one-command deployment for AI agents.
One-Command Deploy
agentd deploy — Sandboxed, autoscaling, production-ready agent hosting.
Deploy your AI agents with zero configuration. We handle infrastructure, scaling, and monitoring.
Plus Hosted Observability
Hosted Dashboard
Visualize runs, traces, and costs. No infrastructure to manage.
Alerts
Get notified when agents fail or costs spike.
Team Collaboration
Shared dashboards, permissions, usage tracking.
Join the Waitlist
Ready to see what your agents are doing?
Join developers building transparent, observable AI agents.
