# Customer-Side Evidence Packet

## Packet

- packet_id:
- contract_version: orbisagent.customer_evidence.v1
- ruleset_id: orbisagent.customer_evidence
- ruleset_version:
- ruleset_hash:
- field_dictionary_version:
- preflight_tool_version:
- packet_date:
- customer_alias:
- prepared_by:
- reviewed_by_human:
- return_channel:
- channel_mode:
- agent_platform:
- agent_source:
- agent_name:
- orbisagent_version_or_template:
- workflow:
- authorized_material_scope:
- permission_scope:
- blocked_actions:
- included_runs:
- evidence_level:
- evidence_files:
- privacy_check:
- preflight_status:
- redaction_method:
- codex_validation_status:
- trust_tier:
- auto_pass_candidate:
- review_sampling_rate:
- structured_payload_format:
- requested_support_actions:
- exception_status:
- status:

## Summary

| Item | Notes |
| --- | --- |
| Actual agent used |  |
| Agent source | orbisagent_9_agent / accio_builtin / customer_owned_agent / third_party_agent / other |
| OrbisAgent version or template |  |
| Deployment object |  |
| Authorized material scope |  |
| Workflows covered |  |
| Runs completed |  |
| Evidence level | L1 / L2 / L3 / L4 |
| Data contract status | passed / failed / partial |
| Preflight status | not_run / passed / failed / partial |
| Ruleset hash |  |
| Codex validation status | not_submitted / accepted / rejected / schema_mismatch / needs_retry |
| Trust tier | t0_mandatory_review / t1_preflight_plus_review / t2_sampled_review / t3_post_review_log_only |
| Review sampling rate |  |
| Outputs adopted |  |
| Outputs partially adopted |  |
| Outputs rejected |  |
| High-risk actions blocked |  |
| Main correction patterns |  |
| Team configuration changes |  |
| Requested support actions |  |

## Run Evidence Cards

| run_id | agent_platform | agent_source | agent_name | ruleset_hash | trust_tier | workflow | output_artifact_type | human_review_result | correction_type | risk_flags | evidence_pointer | status |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| RUN-001 |  |  |  |  | t0_mandatory_review |  | screenshot / markdown_report / sheet_row / manual_note | adopted / partial / rejected / blocked |  |  |  | queued |

## Top Corrections

| correction_id | run_ids | pattern | impact | suggested next action | owner | status |
| --- | --- | --- | --- | --- | --- | --- |
| COR-001 |  |  |  | team config / template update / eval sample / finding candidate / human review boundary |  | queued |

## Risk Gate Events

| event_id | run_id | risk_type | blocked_action | reviewer | result | status |
| --- | --- | --- | --- | --- | --- | --- |
| RISK-001 |  | external message / pricing / backend / budget / contract / payment / privacy / permission |  |  | routed to human | closed |

## Attachments

- run-ledger:
- reviewed output summaries:
- evaluation candidates:
- evidence pointers:

## Privacy Check

- Sensitive full text removed:
- Account secrets removed:
- Signed links removed:
- Backend exports removed:
- Unconfirmed commitments removed:
- Reviewer approved return:

## Exception Handling

- exception_status:
- exception_severity:
- alert_channel:
- dashboard_status:
- retry_due_at:
- expire_at:
- fallback_status:

## Trust Review

- trust_tier:
- auto_pass_candidate:
- minimum_sample_size:
- adoption_rate:
- critical_error_count:
- review_sampling_rate:
- trust_review_due_at:
- trust_revoked_reason: