meta-pdf-intelligence

name	meta-pdf-intelligence
description	Use this meta-skill instead of answering directly when the user needs PDF analysis, pasted PDF excerpt analysis, digesting, comparison, or question answering that benefits from multi-skill orchestration across PDF extraction, summarization, cross-document synthesis, traceable evidence indexing, and memory capture.
kind	meta
meta_priority	55
always	false
final_text_mode	step:cross_document_synthesis
triggers	["看一下这个 PDF","看看这个 PDF","读一下这个 PDF","分析这个 PDF","总结这个 PDF","帮我看 PDF","处理 PDF","PDF 抽要","PDF intelligence","pdf digest","compare these PDFs","page-backed findings","PDF excerpt","pasted PDF excerpt","PDF page says","analyze these PDFs","PDF analysis","PDF comparison","PDF 摘录"]
provenance	{"origin":"opensquilla-original","license":"Apache-2.0"}
composition	{"steps":[{"id":"intake","kind":"llm_chat","with":"[Truncated]"},{"id":"pdf_clarify","kind":"user_input","depends_on":"[Truncated]","when":"'NEEDS_CLARIFICATION: yes' in outputs.intake","clarify":"[Truncated]"},{"id":"extract","skill":"pdf-toolkit","depends_on":"[Truncated]","when":"'SOURCE_STATUS: inline_excerpts_only' not in outputs.intake and (\n 'SOURCE_STATUS: reference_without_content' not in outputs.intake\n or inputs.get('collected', {}).get('pdf_clarify', {}).get('source_status') == 'readable_pdf'\n) and \"don't have the pdf\" not in (inputs.user_message | lower) and \"do not have the pdf\" not in (inputs.user_message | lower) and \"no pdf upload\" not in (inputs.user_message | lower) and \"pdf upload handy\" not in (inputs.user_message | lower) and not ('page ' in (inputs.user_message | lower) and ' says ' in (inputs.user_message | lower))","on_failure":"inline_excerpt_extract","with":"[Truncated]"},{"id":"inline_excerpt_extract","kind":"llm_chat","with":"[Truncated]"},{"id":"per_document_digest","skill":"summarize","depends_on":"[Truncated]","with":"[Truncated]"},{"id":"cross_document_synthesis","kind":"llm_chat","depends_on":"[Truncated]","with":"[Truncated]"},{"id":"traceable_index","kind":"llm_chat","depends_on":"[Truncated]","with":"[Truncated]"},{"id":"memorize","kind":"tool_call","tool":"memory_save","tool_allowlist":"[Truncated]","depends_on":"[Truncated]","tool_args":"[Truncated]"}]}

PDF Intelligence (Meta-Skill)

Process one or more PDFs into a traceable analysis entry. The workflow first classifies the request, preserves file/page evidence, synthesizes across documents when needed, and stores a structured memory index.

Fallback

LLM should manually run pdf-toolkit scripts then summarize and memory_save.