Hypermnesia MCP
Memory for AI coding agents that you can hold accountable. Keep decisions, fixes and project context between sessions, and inspect what was retrieved. It runs locally over stdio via the published package.
People who work with agent memory system, causal inference and cognitive architecture and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in Python.
VERIFIED ACTIVE
LAST COMMIT 2026-09-17 · ★ 73 · #29 OF 304 MAINTAINED MEMORY · VERIFIED 2026-09-18
MIT · Python servers · how we verify → /methodology
01 · Install Hypermnesia MCP
before you install - you'll need
Set CORTEX_MEMORY_STORE_BACKEND before connecting.
Claude Code
claude mcp add cdeust-cortex -- uvx hypermnesia-mcp Claude Desktop / Cursor / VS Code - add to config
{
"mcpServers": {
"cdeust-cortex": {
"command": "uvx",
"args": [
"hypermnesia-mcp"
]
}
}
} Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.
Using another client? Same JSON, different key
Claude Desktop · mcpServers
Cursor · mcpServers
VS Code · servers
Windsurf · mcpServers
Zed · context_servers
Cline · mcpServers
Roo Code · mcpServers
Continue · mcpServers
LibreChat · mcpServers
Gemini CLI · mcpServers
Codex CLI · mcp_servers
Full setup guides: every client.
02 · Evidence
Security posture
What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.
runs as local process (stdio) - runs on your machine with your user's permissions
license MIT - declared in the repository
pypi package hypermnesia-mcp - check the name against the project README before installing (PyPI has no namespace ownership)
registry namespace io.github.cdeust is GitHub-verified and matches the repo owner
03 · What Hypermnesia MCP can do
Prose above is summarized from the project's README and registry record - no invented capabilities.
Latest releases
v4.23.0 · 2026-09-17
4.23.0 (#609) · read Codex rollouts, write the session log before consolidating (#610) (#610) · normalise Codex hook events into what the lifecycle hooks read (#608) (#608) · anchor .Codex/ to repo root, report outside…
v4.22.0 · 2026-09-15
4.22.0 (#570) · remove os/pathlib from ablation.py and 8 retrieval env-flag modules (#560, 2/3) (#568) · remove os/pathlib from wiki_coverage cluster (#560, 3/3) (#569) · remove pathlib from abstention_gate and…
v4.21.0 · 2026-09-10
4.21.0 (#556) · the decision gate judges every language, test file and header (ADR-1060 revision, no exception) (#555) · deliver zetetic-team-subagents v2.41.0 (#554) · deliver zetetic-team-subagents v2.40.1 (#553)…
04 · Who maintains Hypermnesia MCP
hypermnesia-mcp is maintained by cdeust. We track 6 MCP servers from cdeust - 5 actively maintained, 91 combined GitHub stars, oldest repo from Mar 2026. Full record: all servers from cdeust.
- hypermnesia-mcp-viz Read-only visualization for Cortex memory, knowledge, sessions, traces, and codebase graphs. ★ 4
- automatised-pipeline Codebase intelligence MCP server - index any repo into a property graph; 26 tools for agents. ★ 4
- ai-architect-mcp-codebase Codebase intelligence MCP - index any repo into a property graph; 26 tools for agents. ★ 4
- ai-architect-mcp-spec PRD verification and host-driven generation with multi-judge consensus across 17 MCP tools. ★ 3
05 · Facts
- repository
- github.com/cdeust/cortex
- website
- https://ai-architect.tools/
- category
- memory - ranked #29 of 304 actively-maintained memory servers as of 2026-09-18.
- release cadence
- 10+ releases in the last 90 days (latest 2026-09-17)
- registry
- io.github.cdeust/hypermnesia-mcp (active, first published 2026-07-22 · 9 versions)
- packages
- pypi:hypermnesia-mcp
06 · Hypermnesia MCP FAQ
What is Hypermnesia MCP?
Memory for AI coding agents that you can hold accountable. Keep decisions, fixes and project context between sessions, and inspect what was retrieved. It runs locally over stdio via the published package.
Is Hypermnesia MCP still maintained?
Yes - as of 2026-09-18, its last commit was 2026-09-17 and it shipped 10+ releases in the last 90 days. We re-verify nightly.
How do I install Hypermnesia MCP?
Run `uvx hypermnesia-mcp`. The README documents one environment variable (CORTEX_MEMORY_STORE_BACKEND) to set first. Set CORTEX_MEMORY_STORE_BACKEND before connecting. You can also paste the ready-made client config above.
Does Hypermnesia MCP run locally?
Yes - it's a stdio server: it runs on your machine (via uvx) with your user's permissions. Your data stays local unless the server itself calls external APIs.
07 · Alternatives to Hypermnesia MCP
Alternatives to Hypermnesia MCP
Maintained memory servers if Hypermnesia MCP isn't the fit.
- Codebase Memory Codebase knowledge graph for AI agents - 162 languages, sub-ms queries, 99% fewer tokens. ★ 43,715 · 2026-09-18
- Superlog Open-source agent that observes and fixes your application. Query logs, traces, metrics, incidents. ★ 1,449 · 2026-09-17
- Tradememory Protocol Tamper-evident decision audit trail and outcome-weighted memory for AI trading agents. ★ 1,419 · 2026-09-14
- Persome Local-first personal memory and model server for trusted MCP agents on macOS. ★ 1,322 · 2026-08-21
- Neo4j Memory MCP Neo4j Knowledge Graph Memory Server ★ 984 · 2026-09-09
- Projectmem Coding agent memory - one local MCP server for every project. Warns before repeating failed fixes. ★ 826 · 2026-09-15
Pairs well with
Servers that cover what Hypermnesia MCP doesn't - only shown when the pairing reason fits the companion.
More memory MCP servers · Hypermnesia MCP Viz · Remembrall MCP · Memory MCP (chenxiaofie) · Nautilus Compass Memory · Wisdomgraph
More Python MCP servers · Kalshi Prediction Markets · Celmis · Reason Gate · Nable · see all