skills.rest
EricAnalysis
EricAnalysisCOMMUNITY
Explore all skills by this author →
GitHub

eightforge-document-intelligence-reviewer

Reviews document extraction pipelines for provenance, geometry, and verified field integrity.

Updated Mar 8, 2026
Search Tags:#ocr#code review#data integrity#provenance#table extraction#pdf parsing#document extraction
One-click install
npx skills add https://github.com/EricAnalysis/eightforge-os --skill eightforge-document-intelligence-reviewer-ericanalysis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: eightforge-document-intelligence-reviewer
Source: https://github.com/EricAnalysis/eightforge-os/tree/main/.cursor/skills/eightforge-document-intelligence-reviewer
Command: npx skills add https://github.com/EricAnalysis/eightforge-os --skill eightforge-document-intelligence-reviewer-ericanalysis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Document extraction pipelines often fabricate values, lose provenance, or silently drop rows and gaps, making downstream validation and decisions untrustworthy. This Skill provides a structured expert review lens that audits extraction code against strict invariants for immutable source ingest, verified fields, and honest gap reporting. ## Core Features & Use Cases - Provenance and Dependency Auditing: Verifies that every source-derived field closes the chain from immutable source hash through geometry, raw spans, transformations, and normalized values. - Fabrication Detection: Flags authored recovery rows, synthetic anchors, fixed confidence, unlocated AI output, and value-bearing placeholders presented as extracted truth. - Architecture Enforcement: Checks source-neutral extraction, one-way Extraction-to-Interpretation flow, shadow pipeline isolation, and parity/cutover evidence. - Use Case: When reviewing a PR that adds a new XLSX table parser, run this review to confirm table-derived fields carry cell coordinates, rejected candidates are persisted honestly, and no filename-based routing leaked into the generic extraction layer. ## Quick Start Review this pull request touching the PDF extraction pipeline for provenance closure, fabricated values, and honest gap reporting.

Frequently Asked Questions about eightforge-document-intelligence-reviewer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review a document extraction pipeline for data integrity?▼

Review extraction pipelines by tracing each verified field back to its immutable source hash, geometry, raw spans, and replayable transformations. This Skill provides a checklist covering dependency closure, fabrication detection, and honest gap persistence.

What makes a verified field trustworthy in OCR or PDF extraction?▼

A verified field requires an immutable source artifact identity, run and parser identity, valid geometry, ordered source tokens, exact raw reproduction, replayable transformations, and confidence grounded only in that dependency evidence.

Can extraction logic use filenames or vendor names to route parsing?▼

No. Generic extraction must remain source-neutral and may only inspect source bytes, MIME types, observable text, geometry, and layout. Routing by filename, project, vendor, or known expected values belongs in versioned Interpretation mappings, not Extraction.

When should shadow extraction pipelines be promoted to production?▼

Shadow pipelines require dependency closure, genericity and metamorphic evidence, freshness, reviewed differences, preserved production anchors, and an explicit recorded cutover decision. A green parity suite alone never establishes production readiness.

Why do extraction reviews fail on confidence scores?▼

Reviews fail when confidence is fixed, defaulted, or derived from opportunistic value matching rather than dependency evidence. Opaque or absent evidence cannot be upgraded by confidence and must become a rejection, gap, or ambiguity.

Related Skills

Explore More Software Engineering Skills→
ehartyeehartye

clip-docx

Extracts text from Word documents into Markdown clippings for an Obsidian wiki.

Community
Intermediate
PearlThoughtsPearlThoughts
1

extract-requirements

Extract business requirements from undocumented codebases into requirements.md.

Official
Basic
meirmmeirm
29

reverse-engineering-business-logic

Extract business logic from source code into an 11-section structured document.

Community
Advanced
withmartian-sandboxwithmartian-sandbox

ops-platform-onboarding

Standardize service onboarding with Terraform templates, observability, security, and handoff checklists.

Official
Advanced