Trust State
| ID | cmmlx0n5501zdxtikxvg48dzs |
|---|---|
| Agent | agent_charlie |
| Action Type | destructive_delete |
| Reflex Score | 63.9 |
| Autonomy Confidence | 63.8 |
| Prediction Gap | -0.1 |
| Decision | RETRY |
| Status | recorded |
| Tool | — |
| Workflow | wf_finance_update |
| Outcome | — |
| Created | 3/11/2026, 10:47:48 AM |
Signal Breakdown
| Signal | Value | Weight |
|---|---|---|
| Context Integrity | 65.0 | 25% |
| Behavioral Consistency | 62.9 | 20% |
| Tool Risk | 56.4 | 15% |
| Outcome Confidence | 65.4 | 20% |
| Policy Alignment | 67.6 | 20% |
Incident
| Severity | medium |
|---|---|
| Status | open |
| Reason | Repeated retries (17) for workflow wf_finance_update |
Decision Lineage
Counterfactual Simulation
Simulated with: EXECUTE+5, strict destructive block, high tool risk → approval
| Node | Actual | Simulated | Incident? |
|---|---|---|---|
| agent_charlie destructive_delete | RETRY | BLOCK | Yes |
Incident prevention estimate: 1 node(s) would likely have been preventedAdditional approvals/blocks: 0ARR delta: +0.0%GIR delta: +0.0%