memory-ingest — skill
Turns meeting transcripts, chat logs and pasted documents into structured Basic Memory notes. Useful when raw text needs to become entities and relations.
npx skills add basicmachines-co/basic-memory-skills --skill memory-ingest -a claude-codeWhat memory-ingest does
The skill walks unstructured text through a fixed sequence. It parses the format, date and participants, then pulls out entities: people, organizations, topics, action items. Each one gets searched in Basic Memory under several query variations — full name, last name, acronym. Matches are linked with [[wiki-link]]; anything new is proposed for creation.
Nothing is written before you see the list. The skill shows what already exists in memory and what it wants to add, with a line of context per entity, then waits for approval — all at once, one by one, or skip. The source text is kept verbatim, with categorized observations such as opportunity, decision or next_step sitting alongside it.
Web research on new entities is optional and can be skipped when the material already carries enough context. You need Basic Memory itself, with its search_notes and write_note tools — without it there is nothing to search and nowhere to write. Worth checking that the Basic Memory MCP server is connected before you install.
What it helps with
- turn a meeting transcript into a note with participants and decisions
- check memory for existing entities before creating duplicates
- pull commitments and deadlines out of an email thread
- keep a source document verbatim and attach observations to it
- link a new topic to ones already in the graph
Example prompts for the agent
- Process this meeting transcript and add it to Basic Memory
- Go through this email thread — what entities and action items are in it?
- Add this article to Basic Memory as a reference note
How to install memory-ingest
1Skills CLI (Vercel)
npx skills add basicmachines-co/basic-memory-skills --skill memory-ingest -a claude-code2Manually: put the files in place
~/.claude/skills/memory-ingest/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/memory-ingest/
1Skills CLI (Vercel)
npx skills add basicmachines-co/basic-memory-skills --skill memory-ingest -a codex2Manually: put the files in place
.agents/skills/memory-ingest/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add basicmachines-co/basic-memory-skills --skill memory-ingest -a cursor2Manually: put the files in place
.cursor/skills/memory-ingest/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install basicmachines-co/basic-memory-skills memory-ingest2Manually: put the files in place
.github/skills/memory-ingest/SKILL.md1Skills CLI (Vercel)
npx skills add basicmachines-co/basic-memory-skills --skill memory-ingest -a gemini-cli2Platform CLI
gemini skills install https://github.com/basicmachines-co/basic-memory-skills --path memory-ingestManually: put the files in place
.devin/skills/memory-ingest/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add basicmachines-co/basic-memory-skills --skill memory-ingest -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about memory-ingest
How do I install memory-ingest in Claude?
Run npx skills add basicmachines-co/basic-memory-skills --skill memory-ingest -a claude-code. The skill lands at ~/.claude/skills/memory-ingest/SKILL.md.
What does memory-ingest do?
It extracts entities from raw text, cross-references them against Basic Memory, and creates notes with observations and relations once you approve. The original material is preserved verbatim.
What does the skill need to work?
Basic Memory with its search_notes and write_note tools. Without it the skill can neither look up existing entities nor write new notes.