New ID_Check v2.6.11 — a new ID recognition engine improves OCR for Korean documents, PII masking is now set from the dashboard, and submissions can be searched by custom fields (cf1–cf3). See what's new →
New ID_Check v2.6.11 — a new ID recognition engine improves OCR for Korean documents, PII masking is now set from the dashboard, and submissions can be searched by custom fields (cf1–cf3). See what's new →
Expresses how completely the analysis ran as defined, on a 0–100 scale. It is the weighted average of three inputs, and the score band determines a High / Medium / Low level.
Input
Weight
Meaning
Step Execution
40%
Share of playbook-defined steps that actually ran
Step Quality
20%
Step outcomes and whether the configured engines were called
Output Completeness
40%
Share of output schema fields that were populated
The Analysis Score measures output completeness, not the AI’s confidence. Even when every step passes, the score drops if nothing landed in the output schema. The API exposes the same value as confidence.
Decision variables — key values extracted against the output schema, shown as chips. Which items appear depends on your output schema.
Verification steps — the executed steps in order. Each carries a pass indicator plus the rule it applied and the agent’s reasoning.
Recommended actions — follow-up actions generated by the system.
Summary tab — decision variables, verification steps, and recommended actions.
Renders the extracted values as tables that follow the output schema. Section titles and field names come from the output schema you defined, so they differ per workflow.The completeness indicator at the top counts how many top-level schema fields were populated.
Report tab — extracted values grouped by output schema section.
The execution log for every step the agent performed. The header summarizes total steps, passed steps, tool calls, and token usage.Each step shows a category badge (DOCUMENT / IDENTITY / COMPLIANCE / FINANCIAL / OTHER), a result badge (passed / needs_review / failed), token count, and execution time. Expanding Tool reveals the external engine calls.
Audit tab — per-step execution log and external engine calls.
Shows the full extracted data as raw JSON. The Copy button in the top right copies it verbatim, which is handy for integration testing.
Output Schema tab — raw extracted data JSON.
Processing metadata for the analysis.
Processing info — processing time, request time, completion time, playbook ID
Items used — the items this analysis referenced, with their type
Metadata — client metadata passed with the analysis request
Info tab — processing info, items used, and metadata.
