How do I install the Kb Ingest skill?

Run npx skillmds@latest add chendaiwei-99/kb-ingest in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Kb Ingest safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Kb Ingest?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Kb Ingest free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Kb Ingest?

ChenDaiwei-99 (@chendaiwei-99) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Kb Ingest by chendaiwei-99 — Agent Skill

Name: Kb Ingest
Rating: 4.2 (5 reviews)
Author: ChenDaiwei-99

Kb Ingest

Ingest one external source at a time into an Obsidian LLM knowledge base by creating raw-source markdown records for PDFs, literature, GitHub repositories, web pages, or text files. Use when adding raw data to knowledge-base/raw, summarizing a PDF into knowledge-base/raw, documenting a GitHub repo with preserved links and commit context, or updating the LLM wiki after ingestion.

ChenDaiwei-99 Updated Sep 22, 2026

File contents

KB Ingest

Use this skill to turn a raw source into one durable markdown raw-source record. The record should be useful for local search and summary-level reasoning, while preserving original links so future agents can retrieve deeper details online.

Workflow

Read the vault root AGENTS.md and follow any project-specific schema.
Identify the source kind: PDF, GitHub repository, web page, text/markdown file, or other.
Collect metadata, original URLs, local paths, publication dates, authors, repository owner/name, branch, and commit when available.
Inspect the source content carefully enough to write a faithful raw-source record.
Create exactly one markdown record under the matching knowledge-base/raw/ folder using source-record-template.md.
For PDFs, follow pdf-ingest.md.
For GitHub repositories, follow github-ingest.md.
Preserve primary website links in raw-source properties and additional important links in ## Important Details.
Add ## Links Into Wiki entries only for existing or newly created wiki pages.
Update knowledge-base/wiki/index.md and knowledge-base/wiki/log.md directly or invoke the index/log workflow if available.
If contradictions appear, append them to knowledge-base/wiki/review.md; do not rewrite personal notes or contested wiki claims.

Writing Requirements

Write concrete summaries, not promotional blurbs.
Separate Summary, Key Takeaways, Keywords and Concepts, Important Details, and Open Questions.
Keep human-authored prose in personal-notes/ untouched. Useful wikilinks, citation links, backlinks, or frontmatter/property formatting may be added directly; ask for permission before adding or rewriting substantive personal-note content.
Prefer Obsidian wikilinks for local KB pages and normal markdown links for external URLs.

References

source-record-template.md: shared raw-source record template.
pdf-ingest.md: PDF and literature ingestion.
github-ingest.md: GitHub repository ingestion.

---name: kb-ingestdescription: Ingest one external source at a time into an Obsidian LLM knowledge base by creating raw-source markdown records for PDFs, literature, GitHub repositories, web pages, or text files. Use when adding raw data to knowledge-base/raw, summarizing a PDF into knowledge-base/raw, documenting a GitHub repo with preserved links and commit context, or updating the LLM wiki after ingestion.---# KB IngestUse this skill to turn a raw source into one durable markdown raw-source record. The record should be useful for local search and summary-level reasoning, while preserving original links so future agents can retrieve deeper details online.## Workflow1. Read the vault root `AGENTS.md` and follow any project-specific schema.2. Identify the source kind: PDF, GitHub repository, web page, text/markdown file, or other.3. Collect metadata, original URLs, local paths, publication dates, authors, repository owner/name, branch, and commit when available.4. Inspect the source content carefully enough to write a faithful raw-source record.5. Create exactly one markdown record under the matching `knowledge-base/raw/` folder using [source-record-template.md](references/source-record-template.md).6. For PDFs, follow [pdf-ingest.md](references/pdf-ingest.md).7. For GitHub repositories, follow [github-ingest.md](references/github-ingest.md).8. Preserve primary website links in raw-source properties and additional important links in `## Important Details`.9. Add `## Links Into Wiki` entries only for existing or newly created wiki pages.10. Update `knowledge-base/wiki/index.md` and `knowledge-base/wiki/log.md` directly or invoke the index/log workflow if available.11. If contradictions appear, append them to `knowledge-base/wiki/review.md`; do not rewrite personal notes or contested wiki claims.## Writing Requirements- Write concrete summaries, not promotional blurbs.- Separate `Summary`, `Key Takeaways`, `Keywords and Concepts`, `Important Details`, and `Open Questions`.- Keep human-authored prose in `personal-notes/` untouched. Useful wikilinks, citation links, backlinks, or frontmatter/property formatting may be added directly; ask for permission before adding or rewriting substantive personal-note content.- Prefer Obsidian wikilinks for local KB pages and normal markdown links for external URLs.## References- [source-record-template.md](references/source-record-template.md): shared raw-source record template.- [pdf-ingest.md](references/pdf-ingest.md): PDF and literature ingestion.- [github-ingest.md](references/github-ingest.md): GitHub repository ingestion.

Frequently asked questions