research-cataloging

Router for research, archiving, and cataloging: provenance bundles, OCR/transcription, exhaustive source sweeps, translation promotion, catalog reconciliation, NocoDB/Airtable ops, and archival image compression. Use before conversion skills to cover a whole site/course/source set, or whenever you need a transcript. Read one reference. Triggers: Wayback, OCR, coverage ledger, captions, yt-dlp, Whisper.

Jump to install

Repository: KevinGutowski/Skills
Last source activity: July 10, 2026 at 01:01
Detected SKILL.md language: English
Stars: 1
Forks: 0

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

A direct command skips the review prompt. Inspect the source before running it.

npx skills add https://github.com/KevinGutowski/Skills --skill research-cataloging

The command stays on one line. Scroll horizontally to inspect it before copying.

Prefer a local copy? Download the files currently available to SkillsMP.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

File Explorer

22 files

SKILL.md

Source instructions · Read-only preview

Back to search Back to installation

Research & Cataloging (router)

Sources/gaps: references/sources.md maps the research/cataloging references; references/coverage-gaps.md parks source completeness, transcript, database, and handoff standards.

Read exactly one reference file below; each holds the full discipline for its area (deeper worked procedures, helper scripts, and agent manifests live in a same-named subdirectory next to it).

references/archival-research.md — per-subject research bundles: web source trails with snapshots + attributed imagery, OCR/transcription of scanned PDFs into layered corpora, venue/exhibition/artwork dossiers.

references/source-sweep.md — exhaustive coverage of bounded sources: inventory every node, coverage ledger, open every page, never judge from titles.

references/catalog-reconciliation-research.md — making structured records internally consistent: object identities, evidence tiers, image provenance/coverage audits, archive paths, downstream records.

references/source-translation-workflow.md — translation workflows: promoting translations into catalog records, quote fields, bilingual sources.

references/nocodb-catalog-management.md — NocoDB operations: bases/tables/fields, records, linked-record relations, attachments, gallery views, CSV sync.

references/image-archival.md — shrinking oversized images (PNG/TIFF/PSD) to visually identical smaller files: WebP/JPEG/AVIF re-encode, verify lossless, never overwrite originals. (Applies outside research contexts too — any "huge image, compress it" request.)

references/openai-transcription-chunking.md — transcribing audio/video into auditable transcript evidence: caption-first preflight with yt-dlp (YouTube, Instagram, Reels, TikTok, Vimeo, podcasts), then OpenAI/Whisper ASR when captions are missing, with chunking, hallucination audits, and per-chunk retries. Use whenever you need the actual words from a recording — including studying a creator's voice — rather than a webpage's paraphrase.

references/coverage-gaps.md — parked candidate standards for pagination completeness, transcript evidence, database write posture, source-to-skill handoff packets, and scanned-corpus sidecars.

Chains: research → reconciliation → NocoDB sync is the common pipeline; references/archival-research.md is the entry reference and its Related-references block routes onward.

Also here: references/airtable-mcp.md (Airtable MCP setup/auth/troubleshooting in Codex + record ops).