# Icd PDF Analysis

> Use when analyzing ICD/specification PDFs, generating comparison or status documents, or extracting structured tables about message families, frames, subframes, pages, alerts, uplink/downlink flows, RF bands, and missing/open items.

- Skill: `iabuelruzz/icd-pdf-analysis` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add iabuelruzz/icd-pdf-analysis`
- Raw SKILL.md: https://api.skillmd.com/api/skills/iabuelruzz/icd-pdf-analysis/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Docs & Writing
- Author: IAbuElRuzz (https://skillmd.com/u/iabuelruzz)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/iabuelruzz/icd-pdf-analysis

---


# ICD PDF Analysis

Use this skill for ICD/PDF/documentation work.

## Source Handling

- Identify each source document by filename, title, issue/revision, date and role before comparing.
- Extract text with `pdftotext -layout` when building tables or summaries.
- Keep generated deliverables in the user-requested folder.
- Keep source Markdown next to generated PDFs when creating deliverables.
- In this repository, default ICD source PDFs are in `ICD_Documents/`.
- In this repository, default G1/G2 difference, similarity and status outputs are in `ICD_Documents/g1_g2_differences/`.

## Analysis Rules

- Separate confirmed facts from inferences.
- Mark classified, TBD, deleted, spare, reserved, dummy, or not-public items explicitly.
- For page/frame tables, include channel/interface, direction, page or frame size, symbol/bit size, frame/sub-frame timing, interval, fields and source document.
- Call out classified-annex or external-reference limits when public details are not available.
- Do not infer a missing layout or protocol detail unless clearly stated as an inference.

## Output Format

- Prefer Markdown source plus searchable PDF.
- Keep source `.md` next to the generated `.pdf`.
- Use landscape PDF layout for wide comparison tables.
- Include an executive summary, detailed tables, and a missing/open-items register for status documents.

## Useful Commands

- List local ICD PDFs: `find ICD_Documents -maxdepth 2 -type f -name '*.pdf' | sort`
- Extract PDF text: `pdftotext -layout 'path/to/document.pdf' /tmp/document.txt`
- Extract a local ICD PDF: `pdftotext -layout 'ICD_Documents/[AD.xx].pdf' /tmp/adxx.txt`
- Inspect metadata: `pdfinfo 'path/to/document.pdf'`
- Search extracted text: `rg -n "frame|subframe|sub-frame|page|alert|uplink|downlink|band|TBD|classified|reserved" /tmp/document.txt`