Current collected state Current
Observed since 20 Oct, 21:13 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
Tamper-evident execution recording for AI agents — Python SDK
Tamper-evident execution recording for AI agents.
pip install afr-sdk
from afr_python_sdk import Recorder
rec = Recorder(base_url="http://localhost:3001")
run = rec.start({"model": "gpt-4"})
from afr_python_sdk import Recorder
rec = Recorder(base_url="http://localhost:3001", api_key="your-api-key")
rec.start({"model": "gpt-4", "temperature": 0.2})
rec.step({
"step_type": "tool_result",
"tool_name": "risk_score",
"tool_input": {"txn_id": "TXN-001"},
"tool_output": {"fraud_probability": 0.82},
})
rec.decision({
"goal_reference": "evaluate-transaction",
"selected_option": "block",
"confidence": 0.92,
})
proof = rec.finish()
print(proof["run_id"])
MIT
Download counts and popularity will appear here as the collector receives them.
| Version | Published | Direct dependencies | Open version |
|---|---|---|---|
| 1.0.1 | 20 Mar 2026 | 4 |
This describes our collected data; it is not a security assessment.
Observed since 20 Oct, 21:13 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
package:ebfdc08aec2000a4b7ba9475612f5c0679588a21e67d725df18d81b4242ec567