Query Markup Documents - On-device hybrid search for markdown files with BM25, vector search, and LLM reranking
Nothing to run from here
Trove is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
23
tools declared
—
trust score
remote
at its publisher
—
calls last 7d
What it does
Shared memory for your AI agents. Every fact shows its source, and stale facts get replaced.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --transport http anunay999-trove-mcp https://mytrove.in/mcp
# 2 — ask your agent for something these tools do
# remember, recall, grep and 20 more
Trove costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
ANAll 1 server
anunay9991 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. Nothing here runs on your machine and nothing runs on ours: this server answers at its publisher’s own address, and what you install is the connection straight to it.
Claude Codedeclared by this listing
This is a public server: it answers at its publisher’s own address and this marketplace is not in the path. Claude Code registers it in one command; --scope user makes it available in every project. The transport below is http, as the publisher's own manifest declares it. Its publisher asks you to sign in with OAuth: run /mcp in Claude Code and pick this server, and it opens a browser for that once.
claude mcp add --scope user --transport http anunay999-trove-mcp https://mytrove.in/mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it answers at its publisher’s own address: this marketplace is not in the path, so there is no key of ours to hold and no card to add — the only sign-in is its publisher’s own, with OAuth.
What it writes
One config entry pointing straight at the publisher’s own address — this marketplace is not in the path, and nothing from this listing executes on your machine.
What you approve
23 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
Save a short distilled note (a fact, decision, or how-to) — not a raw dump. Same title/slug revises; otherwise creates. Prefer several small linked notes while you work (e.g. 'refunds within 14 days', 'CS owns churn email') over one 'notes from today' blob. Always check returned similar (scored…
No parameter schema on file.
Open questions only — e.g. 'how do we handle refunds?' — returns a short ranked brief with citations. Not for exact ids or error strings (use grep) or when you already know the note name (use read). Default tokenBudget 8000 covers the whole response. Atoms carry the packed body slice —…
No parameter schema on file.
Prefer this over recall for an exact string: ticket id, product code, error text, config key, email, URL path (e.g. INV-1042, ECONNRESET, FEATURE_DARK_MODE). Searches note text and raw sources; invalid regex falls back to literal. Returns excerpts + ids — then read if you need the full note.
No parameter schema on file.
Open one note or raw source by id or name (slug) — full body, not a short brief. For nodes, pass asOf to recover the title, summary, and content recorded at that time. Evidence and annotations remain current. Use when you know the name or after grep/recall found it.
No parameter schema on file.
Link two notes (this decision is for that project, this how-to is part of onboarding). Every new note should hang off a hub. Pass supersedesEdgeId to replace an old link without deleting history.
No parameter schema on file.
Retire a belief that is no longer true. Pass nodeIds/slugs to tombstone whole notes (they leave recall, grep, and read), edgeIds to expire links, or a query to preview first (dryRun defaults true in query mode). Nothing is hard-deleted — supersession history stays queryable via neighborhood…
No parameter schema on file.
List your private notes (newest first, paged by limit/offset; total says how many) so you can propose which ones belong in the team space.
No parameter schema on file.
Move your private notes (nodeIds or slugs) into the team space, where every teammate's agents can recall and read them. Only after the person approved the list. Notes that look like they contain a credential, or whose slug the team already has, are skipped with a reason. The passages a note cites…
No parameter schema on file.
Take shared notes (nodeIds or slugs) out of the team space and back to the private space they were shared from. Only the person who shared a note or a team admin can; notes written straight to the team are skipped.
No parameter schema on file.
Store long raw material (meeting notes, a doc, a paste) as evidence spans. Does not create a findable belief by itself — next: remember 3–7 short facts citing those spans, then connect them to a topic. Pipeline: ingest → remember → connect.
No parameter schema on file.
Tag a source span (supports / contradicts / important quote / …) without creating a new note. Use for provenance marks only.
No parameter schema on file.
Return a bounded graph neighborhood around a node — each node carries its BFS level from the seed. Bound with maxNodes, filter valid-time with validAt, or pass asOf/includeExpired for recorded-time history.
No parameter schema on file.
Render a node as markdown, a mind map, or an agent context pack.
No parameter schema on file.
List saved mind-map and projection views.
No parameter schema on file.
Read a saved mind-map view with included nodes and edges.
No parameter schema on file.
Create a durable saved mind-map view from a root node, search query, or explicit node set.
No parameter schema on file.
Delete a saved mind-map view by id or slug.
No parameter schema on file.
Read cursor-paginated graph mutation events for interface sync.
No parameter schema on file.
No parameter schema on file.
List durable maintenance jobs for projections, lint, and embedding refresh. Scoped to your own jobs; operators without an owner see every job.
No parameter schema on file.
Enqueue a durable maintenance job. Admin scope required.
No parameter schema on file.
Claim and run one pending durable maintenance job inline. Admin scope required.
No parameter schema on file.
How any LLM should use Trove: read routing (grep/read/recall), continuous capture (ingest→remember→connect), supersession, session loop. Read this at session start.
No parameter schema on file.
Free
$0 forever
- Every tool — any call limit is its publisher’s, not ours
- No card, and mcprush charges nothing for it
- Runs at its publisher’s own address — this marketplace is not in the path
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Trove or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
—Grade not published
No rule-by-rule report on file for this release
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
This listing is an address its publisher runs, and you connect to it directly — this marketplace is not in the path. The scanner reads packages and public source, and an address is neither: the repository linked on this page is where this record was collected from, not a copy of what the address serves, so it is not graded in its place. So there is no grade — not a withheld one, an unmeasured one.
Release history
This is an address its publisher runs: which release it serves is theirs to choose, so there is nothing here to pin. The history below is for reading; the current release on file is 0.1.0.
0.1.0
undated
latestNo release note was published with this version.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Trove can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.