
Turns loose documents into structured wiki knowledge. You drop a PDF or text file into your knowledge folder, say "ingest this," and it extracts entities, claims, and relationships, then creates or updates wiki pages with proper cross-references and citations. The workflow is deliberate: it reads the source, discusses key findings with you before writing anything, then updates 5-15 wiki pages per document while maintaining bidirectional wikilinks and logging everything. Good for research projects or technical documentation where you're accumulating knowledge over time and need it interconnected rather than siloed in separate files. The discussion step before writing keeps you in control of what gets emphasized.
$ npx -y skills add open-gitagent/opengap --skill wiki-ingest --agent claude-codeInstalls into .claude/skills of the current project.
Process a new source document and integrate its knowledge into the wiki.
Read the full source document from knowledge/. Identify:
Before writing, share a brief summary of key takeaways. Ask:
For each significant entity or concept found in the source:
memory/wiki/index.mdupdated date and source_countmemory/wiki/ with proper frontmatter, content, and source citations[[wikilinks]] in both directions between related pagesCreate memory/wiki/sources/<source-name>.md with:
memory/wiki/index.md — add/update entries for all pages touchedknowledge/index.yaml — ensure the source is cataloged with tags and prioritymemory/log.md:
## [YYYY-MM-DD] ingest | Source Title
- Pages created: [list]
- Pages updated: [list]
- Key findings: [1-2 sentences]
User drops knowledge/quantum-computing-review-2026.pdf and says "ingest this".
quantum-computing.md, error-correction.md, ibm.md, google.mdsources/quantum-computing-review-2026.mdindex.md with all new/updated pageslog.mdA single source might touch 5-15 wiki pages. This is normal and expected.
Sponsors