What problem does it solve?
This skill solves the problem of manually cross-referencing fragmented public records to uncover hidden relationships between entities, which is extremely time-consuming and error-prone when done by hand across SEC filings, federal contracts, sanctions lists, property records, court documents, and global news archives.
Core Features & Use Cases
- Multi-Source Entity Resolution: Normalize and match entity names across SEC EDGAR, USAspending, OFAC SDN, ICIJ Offshore Leaks, NYC ACRIS property records, OpenCorporates, CourtListener, Wikipedia/Wikidata, and GDELT news using exact, fuzzy, and token-overlap matching with explicit confidence tiers.
- Statistical Timing Correlation: Run permutation tests to determine if event timing clusters suspiciously close together, such as lobbying filings near contract awards, producing p-values and effect sizes to distinguish signal from noise.
- Auditable Evidence Chains: Automatically generate structured findings JSON where every claim traces back to specific source rows with confidence levels, ensuring full auditability and preventing unsupported assertions.
- Use Case: An investigator can trace whether a federal contractor appearing in USAspending also lobbies through Senate LD disclosures, holds NYC property via ACRIS, or appears in OFAC sanctions lists, with all cross-links and timing anomalies documented in a single evidence chain.
Quick Start
Use the osint-investigation skill to investigate "Example Corp" by fetching its SEC EDGAR filings, USAspending contracts, and Senate lobbying records, resolving entities across these sources, and generating a structured findings report with evidence chains.