What problem does it solve? AI-generated answers and reports often contain hallucinated URLs or citations that don't support the claims they reference. This Skill eliminates that risk by having a ledger script own the URL-to-number mapping, so the model only ever emits small integer ids it was handed at retrieval time — a wrong id is detectable and a wrong URL is impossible. ## Core Features & Use Cases - Citation Ledger: Register sources at retrieval time with sources.py add or ingest; ids are stable, idempotent, and URL-normalized across many search rounds. - Cite-While-Drafting Workflow: Place bracketed ids inline per sentence, then mechanically render the Sources block in markdown, plain, footnotes, bibtex, or evidence styles. - Fact-Checking Mode: Attach verbatim quotes to each source (rejected unless they literally appear in the fetched page text), flag model-knowledge claims with [unverified], and gate delivery with verify --evidence --min-coverage. - Use Case: Writing a market research brief from a dozen web sources — register each fetched page, cite claims inline as you draft, then run verify to confirm every citation resolves and coverage meets your threshold before delivering. ## Quick Start Ask the agent to research a topic using web search and write a grounded answer where every factual claim carries an inline numbered citation and a rendered Sources list at the end.
