Turn supplied evidence rows into a citation table with traceable counts

说明与用途

从单一 JSON 包生成可审计的引用表，包含 Markdown、主机计数和来源计数，并提供输入大小限制和审核流程。

问题

Auditing a citation table is hard when Markdown, JSON rows, and summary counts drift apart.

结果

A single, auditable citation table with Markdown, host counts, and source counts that match the input evidence packet.

工作流

• 1. Prepare a JSON packet with schemaVersion 1.0 and up to 100 evidence rows, each containing evidenceId, brandName, eventType, observedAt, sourceName, sourceUrl, and excerpt.
• 2. Submit the JSON to the us‑brand‑evidence‑citation‑table Actor build 0.2.2.
• 3. The Actor returns structured citations, citationMarkdown, hostCounts, and sourceCounts.
• 4. Review the JSON citations and Markdown preview with a reviewer.
• 5. Sample an evidenceId, locate its original record, compare the excerpt, and confirm citation correctness.
• 6. If corrections are needed, update the source packet and regenerate the table.
• 7. Verify the run ID, immutable build ID, and output dataset receipt before using the report.

证据

The useful outcome is a reviewable representation of what you supplied: stable evidence identifiers, attributable excerpts, and counts whose meaning remains clear.

来源

The input is limited to 100 unique evidence IDs and 128 KiB in raw and normalized form.

来源

For a small editorial workflow, give the reviewer both the JSON citations and the Markdown preview.

来源