{"schema_version":"newruntime-agent-readable-v0.2","type":"capability_delta","stable_id":"capability_delta:agents-can-define-document-extraction-schemas-via-natural-language","slug":"agents-can-define-document-extraction-schemas-via-natural-language","title":"Agents can define document extraction schemas via natural language conversation","description":"LlamaIndex’s Retrieval Harness and Conversational Extract let AI agents traverse, query, and extract data from documents without hand‑written JSON schemas, using plain‑language prompts instead.","retrieval_nugget":"LlamaIndex’s Retrieval Harness and Conversational Extract let AI agents traverse, query, and extract data from documents without hand‑written JSON schemas, using plain‑language prompts instead. Before: Agents needed manually authored JSON schemas or code to specify how to parse and extract fields from documents. Enabling change: LlamaIndex released Retrieval Harness for filesystem‑style document tools and Conversational Extract that builds extraction schemas","status":"published","observed_at":"2026-09-26","record_date":"2026-09-26","date_kind":"observed_at","before_capability":"Agents needed manually authored JSON schemas or code to specify how to parse and extract fields from documents.","enabling_change":"LlamaIndex released Retrieval Harness for filesystem‑style document tools and Conversational Extract that builds extraction schemas through dialogue.","after_capability":"Agents can interactively create and run extraction pipelines by describing the desired fields in natural language, enabling on‑the‑fly schema generation.","human_consequence":"Developers and analysts spend far less time authoring schema definitions and can prototype document‑processing flows directly in chat.","verification_level":"source-inspected","buildability_class":"B3","bottleneck":"Accuracy still depends on LLM understanding of the document structure and on the quality of the underlying parsing models.","related_shift":"document-processing-shifts-from-hand-crafted-ocr-pipelines-to-llm","related_posts":[],"topics":[],"source_urls":["https://newruntime.com/signals/parse-extract-classify-now-each-with-its-own-86d89a/","https://newruntime.com/signals/parse-extract-classify-now-each-with-its-own-ce8423/"],"featured":false,"supporting_evidence":[{"title":"Parse, Extract, Classify — now each with its own MCP","url":"https://newruntime.com/signals/parse-extract-classify-now-each-with-its-own-86d89a/","kind":"signal","observed_at":"2026-09-26"},{"title":"Parse, Extract, Classify — now each with its own MCP","url":"https://newruntime.com/signals/parse-extract-classify-now-each-with-its-own-ce8423/","kind":"signal","observed_at":"2026-09-26"}],"routes":{"html":"https://newruntime.com/deltas/agents-can-define-document-extraction-schemas-via-natural-language/","markdown":"https://newruntime.com/deltas/agents-can-define-document-extraction-schemas-via-natural-language.md","json":"https://newruntime.com/deltas/agents-can-define-document-extraction-schemas-via-natural-language.json"},"source_format":"knowledge-ledger","published_at":"2026-09-28T06:03:50.831Z","auto_created":true,"editorial_provenance":{"desk":"knowledge-revision","provider":"groq","model":"openai/gpt-oss-120b","created_at":"2026-09-28T06:03:50.831Z","week_start":"2026-09-21","review":"auto-created from reviewed evidence; the owner may revise or retire it"}}