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
Cogmemai 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
CogmemAi: Autonomous Cognitive Memory for Any Ai System. 95.10% on LongMemEval (top published score on the field's hardest long-term memory benchmark) and 91% on LoCoMo (above human performance). Autonomous memory capture: your Ai's work is saved even whe
Quickstart
# 1 — run it from where its publisher ships it
npx -y cogmemai-mcp
# 2 — the plan is chosen at checkout, in the browser
# 3 — ask your agent something
> CogmemAi: Autonomous Cognitive Memory for Any Ai System. 95.10% on LongMemEval (top published score on the field's hardest long-term memory
Publishernot claimed
HIAll 1 server
hifriendbot1 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. This one runs on your machine: your client starts Cogmemai 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 cogmemai-mcp -- npx -y cogmemai-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 asks of you
It reaches a system of yours, so it needs your own credential rather than the publisher’s: COGMEMAI_API_KEY. You set it after installing; the value is sealed here and sent to the server on each call.
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.
Store a developer memory (fact, preference, decision, architecture detail). Memories persist across all Claude Code sessions and are available in future conversations.
Takes no parameters.
Example callreturns →
Save a mandatory rule that will ALWAYS be followed in every session. Rules bypass scoring and decay — they are injected into every conversation, every time. Use for absolute requirements like
Takes no parameters.
Example callreturns →
List all mandatory rules for the current project and/or globally. Rules are always-on memories that surface in every session.
Takes no parameters.
Example callreturns →
Delete a mandatory rule by its memory ID. Use list_rules to find the ID first.
Takes no parameters.
Example callreturns →
Takes no parameters.
Example callreturns →
Search stored memories using semantic search. Returns memories ranked by relevance, importance, and recency. Use this to find relevant context from past sessions.
Takes no parameters.
Example callreturns →
Extract memories from a conversation exchange using AI. Send the developer message and assistant response, and the server identifies facts worth remembering (architecture decisions, preferences, bug fixes, etc.).
Takes no parameters.
Example callreturns →
Load top memories for the current project plus relevant global memories. Use at the start of a session to get full context from previous sessions. Optionally pass context to get memories most relevant to your current task.
Takes no parameters.
Example callreturns →
List stored memories with optional filters by type, category, scope, or project.
Takes no parameters.
Example callreturns →
Move a memory to the trash by its ID. Trashed memories are hidden from recall and restorable with restore_memory for 30 days, then purged permanently.
Takes no parameters.
Example callreturns →
Restore a trashed (deleted) memory by its ID. Works within 30 days of deletion, before the trash is purged. Use list_trash to find restorable memories.
Takes no parameters.
Example callreturns →
List trashed (deleted) memories, newest first, with days remaining before each is purged permanently. Restore any of them with restore_memory.
Takes no parameters.
Example callreturns →
Update an existing memory
Takes no parameters.
Example callreturns →
Delete multiple memories at once by their IDs. Maximum 100 IDs per call. This is permanent.
Takes no parameters.
Example callreturns →
Update multiple memories at once. Each item needs a memory_id and fields to update. Maximum 50 items per call.
Takes no parameters.
Example callreturns →
Get current usage statistics — memory count, extractions this month, tier info, projects.
Takes no parameters.
Example callreturns →
Export all memories as JSON. Use this to back up memories or transfer them to another project.
Takes no parameters.
Example callreturns →
Bulk import memories from a JSON array. Each memory needs at minimum a content field. Deduplication is applied automatically.
Takes no parameters.
Example callreturns →
Extract memories from a document by splitting it into chunks and processing each one. Great for onboarding — feed in READMEs, architecture docs, or API specs to quickly build project context.
Takes no parameters.
Example callreturns →
Save a summary of the current coding session. Captures what was accomplished, decisions made, and next steps. Stored as a session_summary memory for future reference.
Takes no parameters.
Example callreturns →
List all tags in use across your memories, with counts. Use this to see what threads/groups exist and find related memories by tag.
Takes no parameters.
Example callreturns →
Connect two related memories with a named relationship. Use this to build a knowledge graph — e.g., linking a bug fix to the architecture decision that caused it, or connecting a preference to the pattern it led to.
Takes no parameters.
Example callreturns →
View all memories linked to a specific memory. Returns the relationship type and full memory details for each connection. Use this to explore the knowledge graph around a memory.
Takes no parameters.
Example callreturns →
View the edit history of a memory. Shows all previous versions with timestamps and what changed. Useful for understanding how a decision or fact evolved over time.
Takes no parameters.
Example callreturns →
Get a memory health dashboard with insights: most recalled memories, never-recalled memories, stale memories, growth trends, and breakdowns by type and category. Use this to identify cleanup opportunities and understand memory usage patterns.
Takes no parameters.
Example callreturns →
Promote a project-scoped memory to global scope so it applies across all projects. Use this when you discover a preference or pattern that should be universal — e.g.,
Takes no parameters.
Example callreturns →
Takes no parameters.
Example callreturns →
Create a task that persists across sessions. Tasks are tracked with status (pending, in_progress, done, blocked) and priority (high, medium, low). Use this to maintain continuity on multi-session work.
Takes no parameters.
Example callreturns →
Get tasks for the current project. Returns tasks filtered by status — defaults to showing pending and in_progress tasks. Use at session start to pick up where you left off.
Takes no parameters.
Example callreturns →
Update a task\
Takes no parameters.
Example callreturns →
Save a correction pattern — what went wrong and what the right approach is. These are surfaced automatically when similar situations arise in future sessions, helping avoid repeated mistakes.
Takes no parameters.
Example callreturns →
Set a reminder that surfaces automatically at the start of your next session. Use for follow-ups, things to check, or deferred work. Reminders auto-archive after being shown.
Takes no parameters.
Example callreturns →
Find memories that may be outdated based on age and access patterns. Returns memories that haven\
Takes no parameters.
Example callreturns →
Show what files changed since your last session. Compares the current git state to a snapshot saved when your previous session ended. Helps you understand what happened between sessions.
Takes no parameters.
Example callreturns →
Signal whether a recalled memory was useful or irrelevant. Helps improve future recall quality over time. Use after recalling memories to indicate which were helpful vs noise.
Takes no parameters.
Example callreturns →
Takes no parameters.
Example callreturns →
Analyze memory clusters and extract underlying factual principles or patterns. Principles are observations about your project —
Takes no parameters.
Example callreturns →
Startermost installed
$14 / month
- 11,000 calls a month, included
- 132 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
Pro
$44 / month
- 60,000 calls a month, included
- 720 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
Scale
$94 / month
- 280,000 calls a month, included
- 3,360 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. $14 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
BTrust score 84 / 100
Release 3.23.0 · 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 score91
Client score84
Findings3 threat · 3 capability
Capabilityhigh
Coveragesource
Verificationrepo
Surface digest8f6ffb56576de63b8f141d48a4015071928e6e58771a653f1960c78f2dcf686d
Score vectorthreat 91 → client 84
| Term | Level | What it prices | Points |
|---|---|---|---|
| −6.3 | |||
| −3.15 | |||
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| 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. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| save_memory | no tags | |
| save_rule | no tags | |
| list_rules | no tags | |
| delete_rule | no tags | |
| preflight | no tags | |
| recall_memories | no tags | |
| extract_memories | no tags | |
| get_project_context | no tags | |
| list_memories | no tags | |
| delete_memory | no tags | |
| restore_memory | no tags | |
| list_trash | no tags | |
| update_memory | no tags | |
| bulk_delete | no tags | |
| bulk_update | no tags | |
| get_usage | no tags | |
| export_memories | no tags | |
| import_memories | no tags | |
| ingest_document | no tags | |
| save_session_summary | no tags | |
| list_tags | no tags | |
| link_memories | no tags | |
| get_memory_links | no tags | |
| get_memory_versions | no tags | |
| get_analytics | no tags | |
| promote_memory | no tags | |
| consolidate_memories | no tags | |
| save_task | no tags | |
| get_tasks | no tags | |
| update_task | no tags | |
| save_correction | no tags | |
| set_reminder | no tags | |
| get_stale_memories | no tags | |
| get_file_changes | no tags | |
| feedback_memory | no tags | |
| generate_skills | no tags | |
| extract_principles | 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 digest8f6ffb56576de63b
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 3.23.0 — 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.
3.23.0
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 Cogmemai 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.