# Engram (pypi · engram-chateau)

Local, private long-term memory for AI: verbatim, searchable, correctable. No LLM or API key.

- Trust score: 63/100 (medium)
- Change this week: 0
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- pypi · `engram-chateau`: 63/100 (this document), [markdown](https://verifymcp.io/servers/tushaebxn-engram/engram-chateau.md), [page](https://verifymcp.io/servers/tushaebxn-engram/engram-chateau)

## Channel facts

- Registry: `pypi`
- Package: `engram-chateau`
- Version: `1.10.5`
- 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 2026-11-05.

- **Supply Chain Security**: 75/100
  - No malware found by supply-chain analysis.
  - A critical CVE affects chromadb, and no fixed version has been published yet.
  - Runs setuptools.build_meta at install time, a recognised build step with no custom scripting around it.
  - 8 of 93 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 97/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to TushaeBXN/engram).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 15 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 57/100
  - AI-judged instruction clarity (fair).
  - Tool/resource definitions use about 2026 tokens (~65/item across 31 items; 31 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **Tool Coverage**: 75/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 25% of tool parameters carry a description.
- **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; "engram_delete_drawer" implies "delete" and declares no destructiveHint at all, which the MCP spec reads as destructive by default.
  - An AI judge read all 32 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 20/100
  - Spec-recency check failed: implements MCP spec 2024-11-05; the latest is 2026-07-28.

**Unverified: 1 category.** A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

## Install

### How do I install the Engram MCP server?

Engram runs locally as a PyPI package, launched with uvx engram-chateau. 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 tushaebxn-engram -- uvx engram-chateau
```

### Cursor

```json
{
  "mcpServers": {
    "tushaebxn-engram": {
      "command": "uvx",
      "args": [
        "engram-chateau"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "tushaebxn-engram": {
      "command": "uvx",
      "args": [
        "engram-chateau"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add tushaebxn-engram -- uvx engram-chateau
```

### opencode

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

### OpenClaw

```bash
openclaw mcp add tushaebxn-engram --command uvx --arg engram-chateau
```

### Hermes

```yaml
mcp_servers:
  tushaebxn-engram:
    command: "uvx"
    args: ["engram-chateau"]
```

### Netclaw

```json
{
  "McpServers": {
    "tushaebxn-engram": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "engram-chateau"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add tushaebxn-engram -t stdio -c uvx -a engram-chateau
```

### Other

```json
{
  "mcpServers": {
    "tushaebxn-engram": {
      "command": "uvx",
      "args": [
        "engram-chateau"
      ]
    }
  }
}
```

## 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.

### 2026-10-21 (score 63, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-10-20 (score 48)

First indexed and scored.

## MCP tools (31)

### `engram_search` (~61 tokens)

Semantic search across the château. Returns ranked results with recency weighting.

Input parameters:

- `n` (integer)
- `no_decay` (boolean)
- `query` (string, required): Search query
- `room` (string)
- `wing` (string)

### `engram_add_memory` (~225 tokens)

Store a new memory drawer in the château. The result lists 'possible_updates': existing memories this one may replace. If it does, call engram_supersede (or pass 'supersedes' here).

Input parameters:

- `based_on` (array): IDs of memories that are evidence for this one
- `content` (string, required): Memory content to store
- `derived_from` (array): IDs of memories this one was inferred from
- `hall` (string)
- `pinned` (boolean)
- `reason` (string): Why it replaces it (used with 'supersedes')
- `room` (string)
- `source` (string): Where it came from: user (the person said it), primary (original document, law, ruling, data), secondary (reporting or analysis), tool (search or program output), inference (your own conclusion)
- `source_ref` (string): Citation, URL or document it came from
- `supersedes` (string): ID of a memory this one replaces (optional)
- `wing` (string)

### `engram_wake_up` (~49 tokens)

Return L0+L1 cold-start context (~170 tokens). Load this at the start of every session.

Input parameters:

- `rebuild_l1` (boolean)
- `wing` (string)

### `engram_load_room` (~41 tokens)

Return L2 context for a specific room (ES-compressed closets for all halls).

Input parameters:

- `room` (string, required)
- `wing` (string, required)

### `engram_compress` (~42 tokens)

Compress text into Engram Shorthand (ES).

Input parameters:

- `confidence` (integer)
- `is_code` (boolean)
- `text` (string, required)

### `engram_decompress` (~31 tokens)

Decompress Engram Shorthand (ES) back to natural language.

Input parameters:

- `text` (string, required)

### `engram_kg_query` (~42 tokens)

Query the temporal knowledge graph for an entity.

Input parameters:

- `active_only` (boolean)
- `entity` (string, required)
- `predicate` (string)

### `engram_kg_add` (~63 tokens)

Add a triple to the knowledge graph.

Input parameters:

- `confidence` (number)
- `object` (string, required)
- `predicate` (string, required)
- `subject` (string, required)
- `valid_from` (string)
- `valid_until` (string)

### `engram_kg_invalidate` (~49 tokens)

Invalidate (end) a KG triple.

Input parameters:

- `ended` (string)
- `object` (string, required)
- `predicate` (string, required)
- `subject` (string, required)

### `engram_kg_timeline` (~29 tokens)

Return chronological timeline of all triples for an entity.

Input parameters:

- `entity` (string, required)

### `engram_list_wings` (~18 tokens)

List all wings in the château.

### `engram_list_rooms` (~24 tokens)

List all rooms in a wing.

Input parameters:

- `wing` (string, required)

### `engram_get_drawer` (~47 tokens)

Fetch a specific drawer by ID.

Input parameters:

- `drawer_id` (string, required)
- `hall` (string)
- `room` (string)
- `wing` (string)

### `engram_delete_drawer` (~47 tokens)

Delete a drawer from the château.

Input parameters:

- `drawer_id` (string, required)
- `hall` (string, required)
- `room` (string, required)
- `wing` (string, required)

### `engram_create_wing` (~33 tokens)

Create a new wing in the château.

Input parameters:

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

### `engram_create_room` (~39 tokens)

Create a new room within a wing.

Input parameters:

- `description` (string)
- `name` (string, required)
- `wing` (string, required)

### `engram_diary_write` (~41 tokens)

Write a diary entry for a specialist agent.

Input parameters:

- `agent` (string, required)
- `entry` (string, required)
- `tags` (array)

### `engram_diary_read` (~35 tokens)

Read recent diary entries for a specialist agent.

Input parameters:

- `agent` (string, required)
- `last_n` (integer)

### `engram_status` (~22 tokens)

Return château statistics (wings, rooms, drawers, closets).

### `engram_audit` (~35 tokens)

Run a memory health audit. Returns JSON summary of issues found.

Input parameters:

- `fix` (boolean): Auto-resolve safe issues.

### `engram_replay` (~49 tokens)

Return the chronological story of a room (session replay).

Input parameters:

- `format` (string)
- `limit` (integer)
- `room` (string, required)
- `wing` (string)

### `engram_pin` (~28 tokens)

Pin a drawer so it bypasses recency decay.

Input parameters:

- `drawer_id` (string, required)

### `engram_recall` (~78 tokens)

The few memories clearly relevant to a message, current versions first, or a note that nothing matched. Call it before answering anything that may depend on past work, decisions or preferences.

Input parameters:

- `cwd` (string): Project folder, if known (its memories come first)
- `message` (string, required): The user's message or the task at hand

### `engram_correct` (~172 tokens)

Save a correction: a new memory that replaces an outdated one (same wing/room/hall). The old memory stays searchable but is marked superseded, and search shows the correction first. Safe to retry.

Input parameters:

- `based_on` (array): IDs of memories that support the corrected version
- `content` (string, required): The corrected, current information
- `memory_id` (string, required): ID (or unique prefix) of the outdated memory
- `reason` (string): Why it changed
- `source` (string): Where it came from: user (the person said it), primary (original document, law, ruling, data), secondary (reporting or analysis), tool (search or program output), inference (your own conclusion)
- `source_ref` (string): Citation, URL or document it came from

### `engram_supersede` (~92 tokens)

Record that one stored memory replaces another (e.g. 'we moved to PostgreSQL' replaces 'we use SQLite'). Set undo=true to make the old memory current again. Safe to retry.

Input parameters:

- `new_id` (string): ID of the memory that replaces it
- `old_id` (string, required): ID of the outdated memory
- `reason` (string)
- `undo` (boolean)

### `engram_dispute` (~48 tokens)

Mark two memories as conflicting when it is not yet clear which is right.

Input parameters:

- `id_a` (string, required)
- `id_b` (string, required)
- `reason` (string)

### `engram_history` (~41 tokens)

A memory's status (current, superseded or disputed), what replaced it, what it replaced, when and why.

Input parameters:

- `memory_id` (string, required)

### `engram_link` (~106 tokens)

Record how two stored memories relate: 'memory_id <relation> target_id'. based_on: target is evidence for memory. derived_from: memory was inferred from target. contradicts: they conflict, undecided. supersedes: memory replaces target. undo=true removes the link. Safe to retry.

Input parameters:

- `memory_id` (string, required)
- `reason` (string)
- `relation` (string, required)
- `target_id` (string, required)
- `undo` (boolean)

### `engram_source` (~97 tokens)

Record where a stored memory came from (and optionally the citation or URL).

Input parameters:

- `memory_id` (string, required)
- `source` (string, required): Where it came from: user (the person said it), primary (original document, law, ruling, data), secondary (reporting or analysis), tool (search or program output), inference (your own conclusion)
- `source_ref` (string): Citation, URL or document it came from

### `engram_why` (~100 tokens)

Explain a memory: where it came from, when, whether it is record or inference, the evidence it rests on (and whether that evidence was since corrected or disputed), earlier versions and their evidence, conflicts, and, given the query, why search ranked it. Use it before relying on a conclusion or when asked to justify one.

Input parameters:

- `memory_id` (string, required)
- `query` (string): The question it was retrieved for (optional)

### `engram_check` (~36 tokens)

Memory maintenance report: conclusions whose evidence was corrected or disputed, inferences with no evidence, unresolved conflicts, broken links. Changes nothing.

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Vulnerabilities, Dependencies. The full working is on the page: https://verifymcp.io/servers/tushaebxn-engram/engram-chateau#diagnostics

## Score history

- 2026-11-05: 63
- 2026-11-04: 63
- 2026-11-03: 63
- 2026-11-02: 63
- 2026-11-01: 63
- 2026-10-31: 63
- 2026-10-30: 63
- 2026-10-29: 63
- 2026-10-28: 63
- 2026-10-27: 63
- 2026-10-26: 63
- 2026-10-25: 63
- 2026-10-24: 63
- 2026-10-23: 63
- 2026-10-22: 63
- 2026-10-21: 63
- 2026-10-20: 48

## Common questions

### What is the Engram MCP server?

Engram is an MCP server listed in the public MCP registry as io.github.TushaeBXN/engram. Local, private long-term memory for AI: verbatim, searchable, correctable. No LLM or API key. This page covers its PyPI package (engram-chateau).

### Is the Engram MCP server safe to use?

Engram scores 63 out of 100 on VerifyMCP. We recorded 8 known advisories against it as of 5 November 2026. Its build provenance is signed and verified. 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 Engram MCP server expose?

Engram exposes 31 tools: engram_search, engram_add_memory, engram_wake_up, engram_load_room, engram_compress, and 26 more. Their descriptions and schemas cost roughly 1,820 tokens of context every time the server is loaded.

### Is the Engram MCP server still maintained?

Engram is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the Engram MCP server under?

Engram declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- PyPI project: https://pypi.org/project/engram-chateau/
- Socket report: https://socket.dev/pypi/package/engram-chateau
- Repository: https://github.com/TushaeBXN/engram
- Changelog RSS feed: https://verifymcp.io/servers/tushaebxn-engram/engram-chateau.xml
- Changelog JSON feed: https://verifymcp.io/servers/tushaebxn-engram/engram-chateau.json
- HTML version of this page: https://verifymcp.io/servers/tushaebxn-engram/engram-chateau