# CP Memory (pypi · cp-memory-mcp)

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

- Trust score: 70/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-21

## Components

- pypi · `cp-memory-mcp`: 70/100 (this document), [markdown](https://verifymcp.io/servers/cjhuochai-cp-memory/cp-memory-mcp.md), [page](https://verifymcp.io/servers/cjhuochai-cp-memory/cp-memory-mcp)

## Channel facts

- Registry: `pypi`
- Package: `cp-memory-mcp`
- Version: `1.8.1`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-09-21.

- **Supply Chain Security**: 100/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - Runs setuptools.build_meta at install time, a recognised native-build step with no shell scripting around it.
  - 0 of 29 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 32/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - License check failed: no license is declared.
  - Actively maintained (last published 12 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 65/100
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 1952 tokens (~48/item across 40 items; 40 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 57/100
  - Stability observed for 17 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 71/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "memory_remove" implies "remove" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - An AI judge read all 41 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the CP Memory MCP server?

CP Memory runs locally as a PyPI package, launched with uvx cp-memory-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add cjhuochai-cp-memory -- uvx cp-memory-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "cjhuochai-cp-memory": {
      "command": "uvx",
      "args": [
        "cp-memory-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "cjhuochai-cp-memory": {
      "command": "uvx",
      "args": [
        "cp-memory-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add cjhuochai-cp-memory -- uvx cp-memory-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "cjhuochai-cp-memory": {
      "type": "local",
      "command": [
        "uvx",
        "cp-memory-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add cjhuochai-cp-memory --command uvx --arg cp-memory-mcp
```

### Hermes

```yaml
mcp_servers:
  cjhuochai-cp-memory:
    command: "uvx"
    args: ["cp-memory-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "cjhuochai-cp-memory": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "cp-memory-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add cjhuochai-cp-memory -t stdio -c uvx -a cp-memory-mcp
```

### Other

```json
{
  "mcpServers": {
    "cjhuochai-cp-memory": {
      "command": "uvx",
      "args": [
        "cp-memory-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-09-20 (score 70, +1)

No change was recorded against any check on this day. Stability & Change Management went from 50 to 53. That category is still filling its 30-day observation window: 15 days of observed history at the previous scan, 16 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-18 (score 69, +1)

No change was recorded against any check on this day. Stability & Change Management went from 43 to 47. That category is still filling its 30-day observation window: 13 days of observed history at the previous scan, 14 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-16 (score 68, +1)

No change was recorded against any check on this day. Stability & Change Management went from 37 to 40. That category is still filling its 30-day observation window: 11 days of observed history at the previous scan, 12 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-14 (score 67, +1)

No change was recorded against any check on this day. Stability & Change Management went from 30 to 33. That category is still filling its 30-day observation window: 9 days of observed history at the previous scan, 10 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-11 (score 66, +4)

- [functional improvement] Stability: unverified → 0.23

### 2026-09-07 (score 62, 0)

- [functional regression] Security disclosure: unverified → fail

### 2026-09-06 (score 62, 0)

- [functional regression] Security disclosure: fail → unverified

### 2026-09-05 (score 62, +15)

- [security improvement] Malware scan: unverified → pass

## MCP tools (40)

### `memory_schema` (~20 tokens)

Return CP Memory schema explanations and whether FTS is available.

Output parameters:

- `result` (string)

### `memory_add` (~82 tokens)

Add or update a memory fact. Supports explicit category and optional long-form content payload.

Input parameters:

- `category` (string)
- `confidence` (string)
- `content` (string)
- `content_type` (string)
- `entity` (string, required)
- `property` (string, required)
- `tags` (string)
- `value` (string, required)

Output parameters:

- `result` (string)

### `memory_personal_add` (~132 tokens)

Add or update a long-term personal assistant memory using Profile, Preference, Relationship, Ongoing, Episode, or BeliefDecision.

Input parameters:

- `confidence` (string)
- `details` (string)
- `evidence_count` (integer)
- `key` (string, required)
- `memory_type` (string, required)
- `scope` (string)
- `sensitivity` (string)
- `stability_score` (integer)
- `subject` (string, required)
- `tags` (string)
- `valid_from` (string)
- `valid_until` (string)
- `value` (string, required)

Output parameters:

- `result` (string)

### `memory_personal_derive` (~105 tokens)

Derive a long-term personal memory from an existing Episode and link it back to the source episode.

Input parameters:

- `confidence` (string)
- `details` (string)
- `episode_id` (string, required)
- `evidence_count` (integer)
- `key` (string, required)
- `memory_type` (string, required)
- `stability_score` (integer)
- `subject` (string, required)
- `tags` (string)
- `value` (string, required)

Output parameters:

- `result` (string)

### `memory_episode_consolidate` (~53 tokens)

Preview or apply conservative Episode consolidation into long-term personal memories.

Input parameters:

- `dry_run` (boolean)
- `episode_id` (string, required)
- `limit` (integer)
- `subject` (string)

Output parameters:

- `result` (string)

### `memory_personal_list` (~46 tokens)

List personal assistant memories, optionally filtered by the six-model memory type.

Input parameters:

- `limit` (integer)
- `memory_type` (string)
- `subject` (string)

Output parameters:

- `result` (string)

### `memory_personal_review` (~44 tokens)

Return a user-facing personal memory review dashboard with counts, recent memories, conflicts, and consolidation candidates.

Input parameters:

- `limit` (integer)
- `subject` (string)

Output parameters:

- `result` (string)

### `memory_review_digest` (~48 tokens)

Return a Markdown memory review digest with recent memories, pending auto-extracts, conflicts, stale candidates, and resolution suggestions.

Input parameters:

- `limit` (integer)
- `subject` (string)

Output parameters:

- `result` (string)

### `memory_review_inbox` (~43 tokens)

Show a small actionable inbox for pending memory review. This previews items only and never deletes memory.

Input parameters:

- `limit` (integer)
- `subject` (string)

Output parameters:

- `result` (string)

### `memory_review_apply` (~61 tokens)

Apply one explicit memory review action: confirm, wrong, stale, scoped, or skip. This never physically deletes memory.

Input parameters:

- `action` (string, required)
- `id` (string, required)
- `reason` (string)
- `scope` (string)

Output parameters:

- `result` (string)

### `memory_correct` (~56 tokens)

Mark a memory as corrected, stale, wrong, scoped, or confirmed; optionally replace its value.

Input parameters:

- `id` (string, required)
- `reason` (string)
- `status` (string, required)
- `value` (string)

Output parameters:

- `result` (string)

### `memory_personal_resolve` (~90 tokens)

Resolve a personal-memory conflict by confirming or merging one winner and marking other memories stale, wrong, or scoped with audit links.

Input parameters:

- `loser_ids` (string)
- `loser_status` (string)
- `merged_value` (string)
- `reason` (string)
- `scope` (string)
- `valid_until` (string)
- `winner_id` (string, required)

Output parameters:

- `result` (string)

### `memory_search` (~54 tokens)

Search memories using FTS when available, with OR ranking by default or strict AND matching.

Input parameters:

- `category` (string)
- `limit` (integer)
- `mode` (string)
- `query` (string, required)

Output parameters:

- `result` (string)

### `memory_probe` (~31 tokens)

Get all facts about an entity. Aliases resolve to the canonical entity first.

Input parameters:

- `entity` (string, required)

Output parameters:

- `result` (string)

### `memory_explain` (~43 tokens)

Explain a memory record, including meta, payload, and relations.

Input parameters:

- `entity` (string)
- `id` (string)
- `property` (string)

Output parameters:

- `result` (string)

### `memory_inspect` (~51 tokens)

Inspect a memory record in a user-friendly way, including where the preview, payload, and relations live.

Input parameters:

- `entity` (string)
- `id` (string)
- `property` (string)

Output parameters:

- `result` (string)

### `memory_alias_add` (~40 tokens)

Add or update an alias for a canonical entity.

Input parameters:

- `alias` (string, required)
- `entity` (string, required)
- `tags` (string)

Output parameters:

- `result` (string)

### `memory_alias_list` (~32 tokens)

List aliases, optionally filtered by canonical entity.

Input parameters:

- `entity` (string)
- `limit` (integer)

Output parameters:

- `result` (string)

### `memory_update` (~65 tokens)

Update an existing fact value, confidence, category, or content payload.

Input parameters:

- `category` (string)
- `confidence` (string)
- `content` (string)
- `content_type` (string)
- `id` (string, required)
- `value` (string, required)

Output parameters:

- `result` (string)

### `memory_remove` (~31 tokens)

Delete a fact by ID, including its metadata, payload, and graph links.

Input parameters:

- `id` (string, required)

Output parameters:

- `result` (string)

### `memory_list` (~31 tokens)

List recent facts, optionally filtered by category.

Input parameters:

- `category` (string)
- `limit` (integer)

Output parameters:

- `result` (string)

### `memory_stats` (~26 tokens)

Get memory statistics across facts, payloads, decisions, workflows, aliases, and relations.

Output parameters:

- `result` (string)

### `memory_health` (~24 tokens)

Return CP Memory health, migration state, FTS state, and active task.

Output parameters:

- `result` (string)

### `memory_decision_add` (~53 tokens)

Record an architecture or technical decision and mirror it into facts with strong metadata.

Input parameters:

- `context` (string, required)
- `decision` (string, required)
- `rationale` (string)
- `title` (string, required)

Output parameters:

- `result` (string)

### `memory_decision_list` (~25 tokens)

List recorded decisions, most recent first.

Input parameters:

- `limit` (integer)

Output parameters:

- `result` (string)

### `memory_task_set` (~30 tokens)

Set the current active task.

Input parameters:

- `description` (string)
- `task_name` (string, required)

Output parameters:

- `result` (string)

### `memory_task_get` (~15 tokens)

Get the current active task.

Output parameters:

- `result` (string)

### `memory_task_done` (~27 tokens)

Mark the current task as complete with a result summary.

Input parameters:

- `result` (string)

Output parameters:

- `result` (string)

### `memory_workflow_save` (~38 tokens)

Save a reusable workflow or procedure.

Input parameters:

- `category` (string)
- `name` (string, required)
- `steps` (string, required)

Output parameters:

- `result` (string)

### `memory_workflow_get` (~24 tokens)

Get a saved workflow by name.

Input parameters:

- `name` (string, required)

Output parameters:

- `result` (string)

### `memory_workflow_list` (~21 tokens)

List saved workflows.

Input parameters:

- `category` (string)

Output parameters:

- `result` (string)

### `memory_link_add` (~57 tokens)

Create an explicit memory relation between two records.

Input parameters:

- `relation` (string, required)
- `source_id` (string, required)
- `source_kind` (string, required)
- `target_id` (string, required)
- `target_kind` (string, required)

Output parameters:

- `result` (string)

### `memory_link_list` (~53 tokens)

List explicit memory relations.

Input parameters:

- `relation` (string)
- `source_id` (string)
- `source_kind` (string)
- `target_id` (string)
- `target_kind` (string)

Output parameters:

- `result` (string)

### `memory_touch` (~32 tokens)

Mark matching memories as accessed, optionally by IDs.

Input parameters:

- `ids` (string)
- `query` (string)

Output parameters:

- `result` (string)

### `memory_conflicts` (~29 tokens)

Find possible conflicting memories, including duplicate facts and personal-memory contradictions.

Input parameters:

- `limit` (integer)

Output parameters:

- `result` (string)

### `memory_maintenance` (~49 tokens)

Run memory maintenance: backfill meta, rebuild indexes, and optionally expire low-value records.

Input parameters:

- `dry_run` (boolean)
- `expire` (boolean)
- `limit` (integer)

Output parameters:

- `result` (string)

### `memory_governance_report` (~49 tokens)

Build a non-destructive governance acceptance report over the current memory.db, including auto-extract review queue, conflict samples, corrected samples, and restore probes.

Input parameters:

- `limit` (integer)

Output parameters:

- `result` (string)

### `memory_auto_extract_cleanup` (~73 tokens)

Preview or apply cleanup for auto-extracted noise records that look like implementation explanations rather than real user memory. Prefer action=mark_wrong for safer governance; use delete only when you really want to remove rows.

Input parameters:

- `action` (string)
- `dry_run` (boolean)
- `limit` (integer)

Output parameters:

- `result` (string)

### `memory_restore_context` (~39 tokens)

Build a compact restored context for startup, history recovery, or project continuation.

Input parameters:

- `max_chars` (integer)
- `prompt` (string)

Output parameters:

- `result` (string)

### `memory_recall` (~78 tokens)

CP Memory-first recall入口。任何记忆相关问题都先查 CP Memory 主库，再根据主库命中质量决定是否补查 Codex 自带 memory 作为辅助背景。

Input parameters:

- `allow_auxiliary` (boolean)
- `intent` (string)
- `limit` (integer)
- `query` (string, required)

Output parameters:

- `result` (string)

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/cjhuochai-cp-memory/cp-memory-mcp#diagnostics

## Score history

- 2026-09-21: 70
- 2026-09-20: 70
- 2026-09-19: 69
- 2026-09-18: 69
- 2026-09-17: 68
- 2026-09-16: 68
- 2026-09-15: 67
- 2026-09-14: 67
- 2026-09-13: 66
- 2026-09-12: 66
- 2026-09-11: 66
- 2026-09-10: 62
- 2026-09-09: 62
- 2026-09-08: 62
- 2026-09-07: 62
- 2026-09-06: 62
- 2026-09-05: 62
- 2026-09-04: 47

## Common questions

### What is the CP Memory MCP server?

CP Memory is an MCP server listed in the public MCP registry as io.github.CJhuochai/cp-memory. Local-first, governable long-term memory for AI coding agents. This page covers its PyPI package (cp-memory-mcp).

### Is the CP Memory MCP server safe to use?

CP Memory scores 70 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 21 September 2026. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the CP Memory MCP server expose?

CP Memory exposes 40 tools: memory_schema, memory_add, memory_personal_add, memory_personal_derive, memory_episode_consolidate, and 35 more. Their descriptions and schemas cost roughly 1,900 tokens of context every time the server is loaded.

### Is the CP Memory MCP server still maintained?

CP Memory is still listed as active in the MCP registry. We last reached this channel on 21 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- PyPI project: https://pypi.org/project/cp-memory-mcp/
- Socket report: https://socket.dev/pypi/package/cp-memory-mcp
- Repository: https://github.com/CJhuochai/cp-memory
- Changelog RSS feed: https://verifymcp.io/servers/cjhuochai-cp-memory/cp-memory-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/cjhuochai-cp-memory/cp-memory-mcp.json
- HTML version of this page: https://verifymcp.io/servers/cjhuochai-cp-memory/cp-memory-mcp