Current collected state Current
Observed since 21 Oct, 08:55 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
Security and access control for AI Agents
Security and permission management for AI Agents.
pip install agentwarden
from agentwarden import AgentWarden
# Initialize with your API key
guard = AgentWarden(api_key="your-api-key")
# Check if agent can execute an action
result = guard.check(
agent_id="agent-001",
action="stripe.refund",
context={"amount": 50}
)
if result.allowed:
# Execute your action
stripe.refund.create(amount=50)
# Log the action
guard.log("agent-001", "stripe.refund", "success")
else:
print(f"Action blocked: {result.reason}")
def refund_payment():
return stripe.refund.create(amount=50)
# Check, execute, and log automatically
result = guard.execute(
"agent-001",
"stripe.refund",
refund_payment,
context={"amount": 50}
)
Visit https://agentwarden.io for full documentation.
MIT License
Download counts and popularity will appear here as the collector receives them.
| Version | Published | Direct dependencies | Open version |
|---|---|---|---|
| 0.1.2 | 20 Feb 2026 | 1 |
This describes our collected data; it is not a security assessment.
Shared repository data · GitHub
Observed since 21 Oct, 08:55 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
package:1965e42d8b915dabf8ff63ae51408173c0ba6687fecd9f4bdb621c8cfef3fafa