Memex

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/evan-moon-memex

About

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools

8 read · 5 write · 1 delete

Secrets

None asked for

Install scripts

Not scanned yet

Runs as

Node.js package

Tools (14)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

• delete_note

  Delete a note by ID. Removes both the markdown file and the index entry. Rule-layer notes are user-only and always rejected — suggest `memex delete <id>` instead.

• get_inference

  Get one inference with full provenance: its summary, confidence, and every evidence note (with a snapshot of the note at mint time and whether it has since changed or been deleted). IMPORTANT: an inference is a HYPOTHESIS the engine/agent synthesized, NOT a stored fact. When you use one in an answer, cite it as such with its confidence and evidence note ids (e.g. "[Inference #42, conf 0.78, from notes #1,#2] ..."). Never state it as something the user "knows" or "believes".

• get_note

  Get the full content of a note by ID.

• get_signals

  List deterministic "signals" — un-synthesized patterns the engine found in the notes (NO LLM involved, these are facts about the corpus structure): - hidden_arc: notes that clearly belong together but were never linked, spread over a long time (an un-synthesized thread) - stale_state: a "state" note that has newer related notes since it was last updated - dangling_link: a [[link]] to a note that doesn't exist (an open question) - tag_burst: a tag that went dormant and then resurfaced Use this when the user asks "what patterns/threads/themes are in my notes", "what should I write about", "what…

• list_folders

  List all top-level folders (categories) in the second brain with their note counts. Call this before saving a note to pick the right folder.

• list_inferences

  List inferences (LLM-synthesized hypotheses derived from notes). Staleness is re-checked first, so an inference whose source notes changed shows status "stale". IMPORTANT: an inference is a HYPOTHESIS the engine/agent synthesized, NOT a stored fact. When you use one in an answer, cite it as such with its confidence and evidence note ids (e.g. "[Inference #42, conf 0.78, from notes #1,#2] ..."). Never state it as something the user "knows" or "believes".

• list_notes

  List recent notes from the second brain.

• list_tags

  List all tags in the second brain with their note counts. Call this before saving a note to pick consistent existing tags rather than inventing new ones.

• mint_inference

  Persist a synthesized inference (a non-obvious claim that holds across several notes but is stated in none). ONLY call this when the user EXPLICITLY asks to save/record/keep a discovery ("save that", "record this insight"). NEVER auto-mint in the background — unvalidated hypotheses pollute the brain. First present the hypothesis in chat; mint only after the user approves. Provide either fromSignalId (sources are taken from that signal) or evidenceNoteIds, or both.

• save_note

  Description is built at runtime.

• search_notes

  Search the second brain for relevant context. Call this BEFORE answering any question that could relate to past conversations, people, projects, or decisions the user may have stored. Always search first, then answer — even if the connection seems loose. For important or vague questions, pass MULTIPLE phrasings in one call: once in the user's language and once in English, or once with their wording and once with the underlying concept — results are fused server-side. Short keyword queries work better than long sentences. Returns a compact index (id, title, snippet) where the snippet is the pas…

• set_register

  Record what is currently true about a subject as one value under one key, instead of writing another note that says it. Use this for a fact that gets replaced rather than accumulated — a price, a deadline, a headcount, a monthly figure, a current status. The value is addressed by (subject, predicate, scope), so you never need to know which note held the previous value: writing a new one supersedes it and the old value stays readable as history. Keep writing notes for anything narrative; this is only for a value that has one current answer.

• update_note

  Extend or correct an existing note. Use when new information belongs with an existing note rather than standing alone. Search first to find related notes, then update rather than creating a duplicate. Layer rules: - past notes are immutable — this tool will reject with PAST_IMMUTABLE. Create an [Amendment] save_note with a [[backlink]] instead. - rule notes are user-only — this tool will reject with RULE_USER_ONLY. Surface your proposed change in chat for the user to apply. - state notes update freely. Documents (`operation: "edit-document"`) are a different thing from memory notes. They car…

• update_signal_status

  Triage a signal: dismiss noise so it never resurfaces, or snooze it for later. Use when a signal is clearly not worth synthesizing (e.g. a burst of boilerplate). Confirm with the user before dismissing.

Signals

Stars

3

Forks

0

Open issues

0

Weekly downloads

42

Last commit

Sep 20, 2026 (1 month ago)

Last release

Sep 20, 2026 (1 month ago)

Last package publish

Sep 20, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name

io.github.evan-moon/memex

Version

0.5.0

Auth

No auth

Packages

npm @evan-moon/memex

License

MIT

Repository

github.com/evan-moon/memex

First seen

Oct 24, 2026

Last checked

Oct 25, 2026

Source

Official MCP Registrygithub-graphql