Trace Citations

Trace Citations

Map the citation graph around one focal paper into useful buckets. Use this when the human wants lineage, influence, and strong versus weak citation edges around a paper.

Arguments

• The positional argument is the focal paper query. Quote multi-word titles.
• --depth 1|2 controls whether to expand a second hop from the strongest first-hop edges.
• --max-references <n> and --max-citations <n> cap the first-hop fetch sizes.
• --second-hop-limit <n> caps how many first-hop anchors get expanded at depth two.

Workflow

1. Run python scripts/run.py ....
2. Read result.foundations for strong references behind the focal paper.
3. Read result.direct_descendants for strong citing descendants.
4. Read result.bridge_nodes for medium-confidence connectors with rich context or intent signal.
5. Read result.weak_edges for low-signal edges that are probably less useful.
6. If depth=2, read result.second_hop only after the first-hop picture looks sensible.

Output

• The script prints the unified JSON envelope described in output_contract.md.
• The underlying workflow result is CitationTraceResult.to_dict().
• result.reference_count_examined and result.citation_count_examined show the first-hop search breadth.

When To Escalate

• The focal paper resolves incorrectly.
• The API returns very sparse context and intent data, making edge interpretation weak.
• The first-hop graph is too noisy and needs a tighter focal paper choice before going to depth two.