Agent run audit trail

Research Agent
Session #0a63f2

agent: research-assistant·10 actions·1 scope violation·Feb 23, 2026 · 4h 19m
0
Trust Score
Reliability100
Scope Adherence50
Cost Efficiency85
Latency100
⚠
Agent accessing out of scope resources — api.sec.gov called without authorization. Scope score dropped to 50.

Action Timeline

10 actions · 4h 19m total
cron_triggersystem
success1,444ms689df9c0092e10:31 UTC
api.sec.govhttpSCOPE VIOLATION
200 OK101ms62de5b5b8b9711:44 UTC
write_filetool
error2,573ms2b20a356641311:47 UTC
gpt-4ollm
success3,311msae6063fc367212:57 UTC
api.notion.comhttp
success359ms3a07fb37c11613:31 UTC
claude-sonnet-4-20250514llm
success4,276ms80b789134f5913:35 UTC
read_filetool
success1,309ms3b6262d2a8d114:07 UTC
write_reportfile
success131ms70df95d512b714:14 UTC
read_datasetfile
success165msbed586bc41ea14:17 UTC
summarizetool
error793ms37d8aa4512e914:51 UTC
🔗
Hash Chain Verification
#1
689df9c0
#2
62de5b5b
#3
2b20a356
#4
ae6063fc
#5
3a07fb37
#6
80b78913
#7
3b6262d2
#8
70df95d5
#9
bed586bc
#10
37d8aa45
✓ Hash chain intact — 10 actions verified, no tampering detected

Know what your AI agents
actually did.

Every action. Every tool call. Every decision. Tamper proof audit trails for any AI agent.

Get Early Access Star on GitHub