adeu benchmark report
adeu has a benchmark score of 37/100 based on installability, startup reliability, MCP capability discovery, documentation quality, and safety signals.
Benchmark score
37
Docs score
0
Config score
0
Safety score
0
Benchmark status
| Install | unknown |
|---|---|
| Startup | unknown |
| Capability discovery | passed |
| Last tested | Jul 27, 2026 |
Installability
- npxconfidence: high
npx -y @adeu/mcp-server
Capabilities
resource
ui://adeu/markdown-ui
Adeu Markdown Viewer UI
tool
accept_all_changes
Accepts all tracked changes and removes all comments in a single operation. [Adeu v1.30.0+1fd5285]
tool
diff_docx_files
Compares two DOCX files and returns a compact `@@ Word Patch @@` diff — Adeu's token-level, sub-word patch format — of their text content. Useful for analyzing differences between versions before editing. [Adeu v1.30.0+1fd5285]
tool
finalize_document
Prepares a document for external distribution or e-signature. Note: in this zero-dependency environment, protection_mode='encrypt' is unsupported and falls back to a native read-only lock; export_pdf and password are ignored. [Adeu v1.30.0+1fd5285]
tool
process_document_batch
Applies a batch of edits and review actions to a DOCX. Batches apply SEQUENTIALLY: each change is validated and applied against the document state produced by the changes before it, so a later change may target text an earlier one introduced. Any validation failure rejects the whole batch transactionally — nothing is applied. Each item in `changes` needs a `type`: 1. 'modify': search-and-replace. `target_text` must match uniquely (`match_mode`:'strict', the default) — add surrounding context, or set `match_mode`:'first'/'all'. Set `regex`:true to treat `target_text` as a regex (capture groups in `new_text` as $1, $2…). `new_text` supports Markdown: '#'–'######' headings, '**bold**', '_italic_', '\n\n' paragraph split; empty `new_text` deletes. Never write CriticMarkup ({++, {--, {>>) manually — use the `comment` field. • EMPTY CELLS: a blank table cell has no text to match; `read_docx` renders every cell with a trailing `{#cell:<id>}` anchor — set `target_text` to that exact anchor and put the value in `new_text`. The pipes are display separators, not editable text. 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': add a table row — `target_text` anchors on an existing row's text, `cells` holds the new row's cell values (strings, left to right), `position` is 'above'/'below' (default below). 'delete_row': remove the row matching `target_text`. Disk mode only. ID VOLATILITY: 'Chg:N'/'Com:N' ids shift between document states — always call `read_docx` immediately before accept/reject/reply; never reuse ids from earlier turns. `{#cell:<id>}` anchors are stable across reads and edits, but finalize_document/sanitize regenerates them — re-read after finalizing. `author_name` sets Track Changes attribution; it defaults to 'Adeu AI (TS)' when omitted. [Adeu v1.30.0+1fd5285]
tool
read_docx
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. `page`: a positive integer (1-indexed, default 1) or 'all'. Pages are synthetic length-based chunks sized for LLM consumption, NOT printed Word pages. In mode='full', page='all' returns the whole body with no page chrome. With `search_query`, `page` instead restricts matches to that page (default: search all pages). [Adeu v1.30.0+1fd5285]
Related pages
Frequently asked questions
What does the adeu benchmark score include?
It combines install status, startup status, capability discovery, documentation score, configuration score, and safety score into a 0-100 benchmark_score.
Does MCP Server Finder call adeu tools during discovery?
No. Capability discovery lists tools, resources, and prompts when possible, but does not invoke tools or read sensitive resources.
Why are some checks skipped?
Checks are skipped when they cannot be run safely without credentials, local configuration, or non-destructive execution guarantees.
Last updated