skills.rest
phm-aguiar
phm-aguiarCOMMUNITY
Explore all skills by this author →
GitHub

wiki-ingest

Distills documents, chat exports, images, and URLs into interconnected Obsidian wiki pages.

Updated Aug 2, 2026
Search Tags:#pdf#note-taking#knowledge base#obsidian#wiki#ingest#distillation
One-click install
npx skills add https://github.com/phm-aguiar/test_documentacao --skill wiki-ingest-phm-aguiar
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: wiki-ingest
Source: https://github.com/phm-aguiar/test_documentacao/tree/main/.claude/skills/wiki-ingest
Command: npx skills add https://github.com/phm-aguiar/test_documentacao --skill wiki-ingest-phm-aguiar

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Turning scattered source material — PDFs, chat exports, transcripts, CSVs, images, and web URLs — into a coherent, interlinked knowledge base is tedious and error-prone. This Skill automates the distillation of any source into structured Obsidian wiki pages with wikilinks, provenance tracking, and an ingest manifest. ## Core Features & Use Cases - Multi-format ingestion: Handles markdown, text, PDFs (including academic papers with figure extraction via PyMuPDF), JSON/CSV/HTML data, chat exports (ChatGPT, Slack), images via vision models, and web URLs. - Four ingest modes: Append (hash-based change detection), Full, Raw (promote _raw/ drafts), and Summary (single-page overview for sources over 500KB or 10,000 lines). - Knowledge integration: Creates 10–15 interlinked pages per ingest with typed relationships, confidence scores, provenance markers (^[inferred], ^[ambiguous]), and updates to index.md, log.md, hot.md, and .manifest.json. - Use Case: Drop a ChatGPT conversations.json export into your sources folder and ask the agent to ingest it — it parses the dialogue, extracts reusable knowledge by topic, and writes linked wiki pages while skipping unchanged sources. ## Quick Start Ask the agent to ingest a file or URL into the wiki, for example: "Add this PDF and the notes in my sources folder to the wiki."

Frequently Asked Questions about wiki-ingest

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

FAQPage Schema
How do I ingest documents into an Obsidian wiki automatically?▼

Point the skill at a file, folder, or URL and it reads the source, extracts concepts and claims, then writes interlinked wiki pages with frontmatter, wikilinks, and source attribution. It also updates the vault index, log, and manifest automatically.

How do I process ChatGPT or Slack chat exports into notes?▼

Provide the conversations.json or Slack channel JSON as a source. The skill parses the dialogue structure, skips greetings and meta-conversation, clusters knowledge by topic, and distills it into wiki pages rather than raw transcripts.

Does it skip files that were already ingested?▼

Yes. In append mode it compares each file's SHA-256 hash against the content_hash stored in .manifest.json and skips unchanged sources, re-ingesting only files whose content actually changed.

What happens when a source file is too large to ingest?▼

Files between 500KB–2MB are processed with chunked distillation that merges concepts across chunks. Files over 2MB or 50,000 lines automatically switch to summary mode, producing a single overview page with key claims and a skip report.

Can it extract figures and equations from academic papers?▼

Yes. For arXiv or conference PDFs it re-reads figure-dense pages with vision, extracts embedded images using PyMuPDF, renders vector figures as pixmaps, keeps core equations as LaTeX, and tabulates benchmark results.

Is content from ingested sources treated as instructions?▼

No. Source documents are treated as untrusted data. The skill never executes commands found in sources, never follows embedded instructions, and never makes network requests based on source content.

Related Skills

Explore More Personal & Entrepreneur Skills→
phm-aguiarphm-aguiar

wiki-ingest

Distills documents, chat exports, images, and URLs into interconnected Obsidian wiki pages.

Community
Advanced
Michael1PengMichael1Peng

wiki-query

Answer questions from an Obsidian wiki with citations.

Community
Advanced
Clay-HHKClay-HHK
1

chroma

Store embeddings and metadata in Chroma for vector search and RAG applications.

Community
Intermediate
KemetraKemetra
2

retail-onboard-table

Walks a new raw retail table from Source Ready through Mapping Ready with readiness bookkeeping.

Community
Advanced