Agent Observability for AutoGPT

AutoGPT plugin for structured logging, cost tracking, and compliance audit trails.

Installation

pip install agent-observability-autogpt

Setup

1. Get Your Free API Key

curl -X POST https://api-production-0c55.up.railway.app/v1/register \
  -H "Content-Type: application/json" \
  -d '{"agent_id":"my-autogpt"}'

2. Set Environment Variable

export AGENT_OBS_API_KEY=ao_live_your_key_here

3. Enable in AutoGPT

Add to your AutoGPT plugins configuration:

agent-observability-autogpt

What Gets Logged Automatically

Manual Logging

Your AutoGPT agent can also log custom events:

from agent_observability_autogpt import AgentObservabilityPlugin

plugin = AgentObservabilityPlugin()

# Log an API call with cost
plugin.log_event(
    event_type="api_call",
    metadata={
        "provider": "openai",
        "model": "gpt-4",
        "cost_usd": 0.03,
        "latency_ms": 1200
    }
)

# Log a decision
plugin.log_event(
    event_type="decision",
    metadata={
        "decision": "use_web_search",
        "reason": "need current information"
    }
)

# Get task summary
summary = plugin.get_task_summary()
print(f"Task cost so far: ${summary['total_cost_usd']}")
print(f"Commands executed: {summary['commands_executed']}")

Features

• Automatic Logging: All commands and responses logged automatically
• Cost Tracking: Track spending per task
• Task Metrics: Elapsed time, command count, cumulative costs
• Compliance Ready: SOC 2, GDPR audit trail requirements
• Zero Configuration: Just set the API key

Pricing

Related Packages

• agent-observability — Core Python SDK
• agent-observability-langchain — LangChain integration
• agent-observability-crewai — CrewAI integration

License

MIT License — see LICENSE for details.

Links

• API Documentation