Automate ingestion of project specs, agent logs, and wiki pages into Minni memory.
npx skills add https://github.com/infektyd/minni --skill minni-ingestion
Please help me install this Agent Skill. Skill: minni-ingestion Source: https://github.com/infektyd/minni/tree/main/plugins/minni/skills/minni-ingestion Command: npx skills add https://github.com/infektyd/minni --skill minni-ingestion
Content from project specs, agent conversation logs, identity files, and LLM wikis remains scattered and unindexed in Minni's memory system, requiring manual work to ingest and leading to missing context for AI agents.
Use the minni-ingestion skill to automatically scan your recent project docs, agent conversations, and LLM wiki pages and ingest them into your Minni memory vault with enriched metadata and fact extraction.
High-intent search queries and answers about installing and using this skill.
You can automate wiki page ingestion by using scheduled scans of recently modified files, extracting facts on-device, and enriching chunk headings with frontmatter metadata. This process ensures reliable, low-noise memory population without manual indexing.
Resolving wikilinks into knowledge graph edges is handled through a two-phase wikilink resolution mechanism during wiki page ingestion. This approach maps [[wikilinks]] into structured edges while adding source provenance tagging for filtered recall.
Auto-indexing handles modified files by scanning them every 30 minutes and extracting structured facts using on-device Apple Foundation Models. Extracted facts are written to an auto-indexed folder for immediate pickup by the memory system.
Yes, ingestion implements timestamp-aware deduplication to prevent duplicate memory entries. It also features bulk copy exclusion and large file timeout handling to ensure reliable content processing without system stalls.
Yes, you can ingest scattered content from project specs, agent conversation logs, and identity files automatically. The ingestion paths eliminate manual entry by pulling new documentation directly into the shared memory store for up-to-date agent context.
Lucius646Manage Markdown-based knowledge bases with LLM-driven ingestion, querying, and maintenance.
zocomputerSave facts, decisions, and preferences into a knowledge graph via supermemory.ai API.
jtvarisIntegrate on-device Foundation Models with session management and structured outputs.
ShaheerKhawajaRetrieve wiki vault pages via progressive loading and wikilink traversal for cross-domain queries.
