io.github.CJhuochai/cp-memory

CP Memory

Is the CP Memory MCP server down? No. It answered initialize and tools/list in the probe of 2026-11-01.

Local-first, governable long-term memory for AI coding agents

alive probed 2026-11-14 · unchanged since 2026-11-01

40tools

6.8ktokens in Claude

3.4 %of Claude's 200k window

2.7×the median server

2.1 sto initialize

7 msto list tools

Run it: uvx cp-memory-mcp@1.8.1 (stdio, probed in a container with the network off)
Host: pypi.org
Reports itself as: cp-memory 1.30.0
Capabilities: experimental, prompts, resources, tools
Instructions: 223 characters
Of a 1M window: 0.7 % of a 1M context window
Count: 6.8k tokens in Claude, exact, measured 2026-10-25. With the Claude Code prefix mcp__…__ on each tool name: 7.2k. The o200k_base estimate is 3.5k. Counted again when the tool definitions change. Method.
Repository: https://github.com/CJhuochai/cp-memory
Registry: active, updated 2026-09-04 · registry entry
Changes: 2 recorded, reachable in 2 of them. A row is added when the result, the tool count or the cost changes.

Most expensive tools

The Claude column scales the o200k estimate of each tool by this server's ratio of exact to estimate, 1.96×.

Tool	≈ Claude tokens	o200k estimate	Share of this server
memory_personal_add	489	249	7 %
memory_personal_derive	396	202	6 %
memory_personal_resolve	328	167	5 %
memory_add	306	156	4 %
memory_recall	249	127	4 %

Other registry entries on pypi.org, the ones next to this name; all of them.

Servers that mention memory.

We add a row when the result changes. A server with no change keeps its last row.

Probed	Result	Init	Tools	Claude	o200k
2026-10-25 06:52	alive	2.9 s	40	6.8k	3.5k
2026-10-18 06:39	alive	2.1 s	40		3.5k

The result of the latest probe, the tool count and the token cost. The feed has an entry for every probe that changed them.

Stops answering, comes back, tool set or cost changes. One confirmation mail first, a stop link in every mail.