MCP server for 0Latency — persistent memory layer for AI agents. Give Claude, Cursor, or any MCP client long-term memory.
Free
Nothing to run from here
Agent Recall 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.
Open Installation—
p95 latency
—
call success
local
runs on your machine
—
calls last 7d
What it does
Memory Palace MCP server for AI agents
Quickstart
# 1 — run it from where its publisher ships it
npx -y agent-recall-mcp
# 2 — the plan is chosen at checkout, in the browser
# 3 — ask your agent something
> Memory Palace MCP server for AI agents
Publishernot claimed
GOAll 3 servers
goldentrii3 servers
2 categories
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. This one runs on your machine: your client starts Agent Recall as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. It is scanned, signed and pinned to the version you choose — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. Claude Code registers it in one command.
claude mcp add agent-recall-mcp -- npx -y agent-recall-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
This is a paid listing, so the CLI has to be signed in and the licence has to be on your account — how the login works. Until then the install stops with the price rather than half-writing a config entry.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
37 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.
[ENTRY — call FIRST, before acting] Load project context at session start — corrections, insights, watch_for warnings
Takes no parameters.
Example callreturns →
Pre-action safety matcher — warns on publish/push/deploy (--full)
Takes no parameters.
Example callreturns →
Open a new project narrative phase (AR_EXTRAS=1 --full)
Takes no parameters.
Example callreturns →
Record what the agent understood, its confidence, and any human correction. Measures the Intelligent Distance gap.
Takes no parameters.
Example callreturns →
Add insights to the awareness system. Call at end of session.
Takes no parameters.
Example callreturns →
[MID-SESSION — safe any time; for alignment, before risky decisions] Use when the user asks to validate understanding, verify alignment, or check if their interpretation matches the human
Takes no parameters.
Example callreturns →
Generate L3 semantic synthesis from recent journals and palace rooms.
Takes no parameters.
Example callreturns →
Context cache — store or recall pre-computed analysis results.
Takes no parameters.
Example callreturns →
Move entries older than N days to cold archive. Keeps a one-line summary per archived entry.
Takes no parameters.
Example callreturns →
Layer 1: lightweight Q&A capture. Appends to today
Takes no parameters.
Example callreturns →
Returns a palace-first cold-start package.
Takes no parameters.
Example callreturns →
List available journal entries for a project.
Takes no parameters.
Example callreturns →
List all projects tracked by agent-recall on this machine.
Takes no parameters.
Example callreturns →
Read a journal entry. Returns the full file content for agent cold-start. Use date=
Takes no parameters.
Example callreturns →
Condense old daily journals into weekly summaries.
Takes no parameters.
Example callreturns →
Full-text search across all journal entries for a project.
Takes no parameters.
Example callreturns →
Layer 1: structured JSON session state. Faster than markdown for cold-start.
Takes no parameters.
Example callreturns →
Append content to the current journal entry (creates today
Takes no parameters.
Example callreturns →
Read lessons from knowledge files. Used before starting work to learn from past mistakes.
Takes no parameters.
Example callreturns →
Write a structured lesson to the knowledge palace room.
Takes no parameters.
Example callreturns →
Surface a contradiction between the human
Takes no parameters.
Example callreturns →
Health check: find stale memories, orphan rooms (no connections), low-salience entries,
Takes no parameters.
Example callreturns →
Read a room overview or specific topic from the Memory Palace.
Takes no parameters.
Example callreturns →
Full-text search across all palace rooms. Results are ranked by room salience.
Takes no parameters.
Example callreturns →
Progressive context loading for cold-start.
Takes no parameters.
Example callreturns →
Write a memory to a palace room. Triggers fan-out: cross-references are updated in connected rooms.
Takes no parameters.
Example callreturns →
Close the currently active project phase with the three reflection fields:
Takes no parameters.
Example callreturns →
Return the full content of the currently active project phase, or null if none is open.
Takes no parameters.
Example callreturns →
List all project phases (milestones) in order — the project
Takes no parameters.
Example callreturns →
Render the project
Takes no parameters.
Example callreturns →
[RETRIEVE — use freely, any time] Use when the user asks to recall, search, find, or look up previous memory, context, or decisions.
Takes no parameters.
Example callreturns →
Before starting a task, recall cross-project insights that apply.
Takes no parameters.
Example callreturns →
Save a permanent IF-THEN behavior commitment for the project. Unlike insights (semantic facts)
Takes no parameters.
Example callreturns →
[MID-SESSION WRITE — single fact/decision; saying it is not saving it] Use when the user asks to remember, store, note, or save a specific decision, fact, or insight.
Takes no parameters.
Example callreturns →
[ON SAVE/EXIT — YOU must call this; nothing auto-saves] Use when the user asks to save, checkpoint, summarize, end, retain, or persist the current session.
Takes no parameters.
Example callreturns →
Search ALL memory stores at once — palace, journal, and insights.
Takes no parameters.
Example callreturns →
Save a memory — the system classifies your content and routes it to the right store
Takes no parameters.
Example callreturns →
Startermost installed
$10 / month
- 8,000 calls a month, included
- 96 calls a minute at peak
- Every tool the listing has
- Cancel any time — the unused days are credited
- Runs behind the gateway, so cancelling stops it
Team
$35 / month
- 40,000 calls a month, included
- 480 calls a minute at peak
- Every tool the listing has
- Cancel any time — the unused days are credited
- Runs behind the gateway, so cancelling stops it
Business
$75 / month
- 180,000 calls a month, included
- 2,160 calls a minute at peak
- Every tool the listing has
- Cancel any time — the unused days are credited
- Runs behind the gateway, so cancelling stops it
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
No call is billed on its own, and this plan carries no monthly call limit. $10 a month covers every tool 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: $8.42
$16.58 left
ATrust score 96 / 100
Release 3.4.46 · scanned 25 Aug 2026
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.
Threat score100
Client score96
Findings0 threat · 0 capability
Capabilitymoderate
Coveragesource
Verificationrepo
Surface digest3ca06910017fb5e4b38eda9f0dba1cbe09c8dbbea5c507a10a5529436e362f12
Score vectorthreat 100 → client 96
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | moderate | capability blast radius (moderate) — client exposure if the model is manipulated | −3 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (37 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. moderate is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| alignment_check | no tags | |
| awareness_update | no tags | |
| check_action | no tags | |
| check | external-sink | |
| context_synthesize | no tags | |
| digest | no tags | |
| journal_archive | no tags | |
| journal_capture | no tags | |
| journal_cold_start | no tags | |
| journal_list | no tags | |
| journal_projects | no tags | |
| journal_read | no tags | |
| journal_rollup | no tags | |
| journal_search | no tags | |
| journal_state | no tags | |
| journal_write | no tags | |
| knowledge_read | no tags | |
| knowledge_write | no tags | |
| nudge | no tags | |
| palace_lint | no tags | |
| palace_read | no tags | |
| palace_search | no tags | |
| palace_walk | no tags | |
| palace_write | no tags | |
| pipeline_close | no tags | |
| pipeline_current | no tags | |
| pipeline_list | no tags | |
| pipeline_open | no tags | |
| pipeline_show | no tags | |
| recall_insight | no tags | |
| recall | no tags | |
| register_rule | no tags | |
| remember | no tags | |
| session_end | no tags | |
| session_start | no tags | |
| smart_recall | no tags | |
| smart_remember | no tags |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digest3ca06910017fb5e4
Scanned2026-08-25
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Pinned to 3.4.46 — the install command below asks for that release. A pin is part of an install, so it is kept for this visit and written down when you install.
3.4.46
undated
installedNo release note was published with this version.
—
no reviews yet
Reviews are earned
Only accounts with at least 50 real tool calls against this server in the last 90 days can post. Ratings are weighted by how much the reviewer actually uses it, and publishers can reply once per review.
Reviews are earned
Writing one takes an account with at least 50 real tool calls against Agent Recall in the last 90 days. That is the whole gate — there is no other way to post, which is why the counts beside each review are worth reading.
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 an account with 50 real tool calls against it.