# ai.adeu/adeu (npm · @adeu/mcp-server)

Automated DOCX Redlining Engine

- Trust score: 76/100 (medium)
- Change this week: +33
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- npm · `@adeu/mcp-server`: 76/100 (this document), [markdown](https://verifymcp.io/servers/ai-adeu-adeu/adeu-mcp-server.md), [page](https://verifymcp.io/servers/ai-adeu-adeu/adeu-mcp-server)

## Channel facts

- Registry: `npm`
- Package: `@adeu/mcp-server`
- Version: `1.7.1`
- 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 2026-08-03.

- **Supply Chain Security**: 87/100
  - No malware found by supply-chain analysis.
  - Only part of the dependency tree could be resolved (99 of 103), so this covers what we could see, not the whole tree.
  - No install/post-install scripts declared.
  - Only part of the dependency tree could be resolved (99 of 103), so this covers what we could see, not the whole tree.
- **Provenance & Transparency**: 97/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to dealfluence/adeu).
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 9 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 63/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 1318 tokens (~146/item across 9 items; 9 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 27/100
  - Stability observed for 8 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 86/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 59% of tool parameters carry a description.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add ai-adeu-adeu -- npx -y @adeu/mcp-server
```

### Codex

```bash
codex mcp add ai-adeu-adeu -- npx -y @adeu/mcp-server
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ai-adeu-adeu": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "@adeu/mcp-server"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ai-adeu-adeu --command npx --arg -y --arg @adeu/mcp-server
```

### Hermes

```yaml
mcp_servers:
  ai-adeu-adeu:
    command: "npx"
    args: ["-y", "@adeu/mcp-server"]
```

### Other

```json
{
  "mcpServers": {
    "ai-adeu-adeu": {
      "command": "npx",
      "args": [
        "-y",
        "@adeu/mcp-server"
      ]
    }
  }
}
```

## 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.

### 2026-08-03 (score 76, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-08-02 (score 75, +42)

- [security improvement] Install scripts: unverified → pass
- [security improvement] Provenance: unverified → pass
- [security improvement] Known CVEs: unverified → partial
- [security] The attested source repository moved: dealfluence/adeu
- [functional regression] Security disclosure: fail → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional improvement] License: unverified → pass
- [functional improvement] Dependency health: unverified → partial
- [functional improvement] Maintenance: unverified → pass
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Stability: unverified → 0.23
- [functional improvement] Schema quality: unverified → good
- [functional] Licence: MIT

### 2026-08-01 (score 33, +8)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-31 (score 25, −18)

- [security regression] Malware scan: pass → unverified

### 2026-07-29 (score 43, +19)

- [functional improvement] Tool coverage: unverified → 100

### 2026-07-28 (score 24, −19)

- [functional regression] Tool coverage: 100 → unverified
- [functional] First check of Schema quality: unverified

### 2026-07-27 (score 43)

First indexed and scored.

## MCP tools (9)

### `read_docx` (~288 tokens)

Reads a DOCX file. Returns text with inline CriticMarkup for Tracked Changes and Comments: {++inserted++}, {--deleted--}, {==highlighted==}{>>comment<<}. Set clean_view=True for the finalized 'Accepted' text without markup.

Modes:
\- 'full' (default): paginated body content. Use page=N to navigate.
\- 'outline': heading map only — start here for large docs to plan targeted reads. Defaults to L1-L2 headings; pass outline_max_level=3-6 to see deeper structure.
\- 'appendix': defined terms, anchors, and cross-reference targets. Consult before editing legal/technical docs to avoid breaking references.

Input parameters:

- `clean_view` (boolean): If False (default), returns the 'Raw' text with inline CriticMarkup. If True, returns 'Accepted' text.
- `file_path` (string, required): Absolute path to the DOCX file.
- `mode` (string): 'full' returns body content. 'outline' returns a structural heading map. 'appendix' returns defined terms.
- `outline_max_level` (number): For mode='outline' only: cap on heading depth.
- `outline_verbose` (boolean): For mode='outline' only: includes metadata.
- `page` (number): Page number (1-indexed) for mode='full'. Defaults to 1.

### `process_document_batch` (~391 tokens)

Applies a batch of edits and review actions to a DOCX.

All changes evaluate against the ORIGINAL document state — do not chain dependent edits within one batch (e.g. rename X to Y, then modify Y). Apply the rename first, then send a second batch.

Each item in `changes` must specify a `type`:
1\. 'modify': Search-and-replace. `target_text` must uniquely match — include surrounding context if the phrase is ambiguous. `new_text` supports Markdown: '# Heading 1' through '###### Heading 6', '**bold**', '_italic_', and '\n\n' to split into multiple paragraphs. Empty `new_text` deletes. Do NOT write CriticMarkup tags ({++, {--, {>>) manually — use the `comment` parameter for comments.
2\. 'accept' / 'reject': Finalize or revert a tracked change by `target_id` (e.g. 'Chg:12').
3\. 'reply': Reply to a comment by `target_id` (e.g. 'Com:5') with `text`.
4\. 'insert_row' / 'delete_row': Table edits. Disk mode only — not supported on Live Word canvas.

ID VOLATILITY: 'Chg:N' and 'Com:N' shift between document states. Always call `read_docx` immediately before any accept/reject/reply — do not reuse IDs from earlier in the conversation.

\`author_name` is used for attribution on all tracked changes and comments, in both disk and Live Word modes.

Input parameters:

- `author_name` (string, required): Name to appear in Track Changes (e.g., 'Reviewer AI').
- `changes` (array, required): List of changes to apply. Each change must specify 'type'.
- `original_docx_path` (string, required): Absolute path to the source file.
- `output_path` (string): Optional output path.

### `accept_all_changes` (~74 tokens)

Accepts all tracked changes and removes all comments in a single operation, producing a finalized clean document. Use this when a document review is entirely complete and you want to clear all redlines.

Input parameters:

- `docx_path` (string, required): Absolute path to the DOCX file.
- `output_path` (string): Optional output path.

### `diff_docx_files` (~90 tokens)

Compares two DOCX files and returns a unified diff of their text content. Useful for analyzing differences between versions before editing.

Input parameters:

- `compare_clean` (boolean): If True, compares 'Accepted' state. If False, compares raw text.
- `modified_path` (string, required): Absolute path to the modified DOCX file.
- `original_path` (string, required): Absolute path to the baseline DOCX file.

### `finalize_document` (~190 tokens)

Prepares a document for external distribution or e-signature. This tool combines metadata sanitization, document locking (protection), and markup resolution into a single step. NOTE: PDF export and AES encryption are disabled in this environment.

Input parameters:

- `accept_all` (boolean): If true, auto-accepts all unresolved track changes before finalizing.
- `author` (string): Replace all remaining markup authorship with this name.
- `export_pdf` (boolean): Ignored in this environment.
- `file_path` (string, required): Absolute path to the DOCX file.
- `output_path` (string): Optional output path.
- `password` (string): Ignored in this environment.
- `protection_mode` (string): Native OOXML document locking. encrypt falls back to read_only in this environment.
- `sanitize_mode` (string): full removes all markup, keep-markup redacts metadata but keeps comments/redlines.

### `login_to_adeu_cloud` (~31 tokens)

Logs the user into the Adeu Cloud backend. Securely opens a browser window for authentication.

### `logout_of_adeu_cloud` (~27 tokens)

Logs out of the Adeu Cloud backend by clearing the local API key.

### `search_and_fetch_emails` (~136 tokens)

Searches the user's live email inbox. By default, searches only the Inbox folder. Returns a list of lightweight previews. Call again with `email_id` to fetch the full body and download attachments.

Input parameters:

- `attachment_name` (string)
- `days_ago` (number)
- `email_id` (string)
- `folder` (string)
- `has_attachments` (boolean)
- `is_unread` (boolean)
- `limit` (number)
- `offset` (number)
- `sender` (string)
- `subject` (string)
- `working_directory` (string)

### `create_email_draft` (~91 tokens)

Creates an email draft in the user's native draft box. Provide `reply_to_email_id` to reply, or `subject` and `to_recipients` for a new email.

Input parameters:

- `attachment_paths` (array)
- `body_markdown` (string, required)
- `reply_to_email_id` (string)
- `subject` (string)
- `to_recipients` (array)

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/ai-adeu-adeu/adeu-mcp-server#diagnostics

## Score history

- 2026-08-03: 76
- 2026-08-02: 75
- 2026-08-01: 33
- 2026-07-31: 25
- 2026-07-29: 43
- 2026-07-28: 24
- 2026-07-27: 43

## Links

- npm package: https://www.npmjs.com/package/@adeu/mcp-server
- Socket report: https://socket.dev/npm/package/@adeu/mcp-server
- Repository: https://github.com/dealfluence/adeu
- Changelog RSS feed: https://verifymcp.io/servers/ai-adeu-adeu/adeu-mcp-server/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/ai-adeu-adeu/adeu-mcp-server/changelog.json
- HTML version of this page: https://verifymcp.io/servers/ai-adeu-adeu/adeu-mcp-server