Anno MCP server
io.github.philmingdao/anno·v0.4.0·1 starsLocal-first HTML review, editing, and annotation for AI coding agents.
1 stars57 downloads/wk
Reviews
Write oneNobody has reviewed Anno yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Anno tools (6, 1 write)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
html_review_claim_handoffThe receiving agent turn must call this first to provide a durable receipt that the exact Anno generation request reached the current session. Idempotent for the active handoff.
html_review_get_sessionRead the current session status, output paths, edit counts, and complete annotation payload. Use after the user finishes browser review or when resolving pending free-form annotations.
html_review_list_sessionsList recent local HTML review sessions with concise status and output metadata.
html_review_mark_handoff_sentRecord that the host accepted an Anno follow-up message for transport. This is not an agent receipt; the receiving turn must still call html_review_claim_handoff.
html_review_register_finalRegister a fully resolved HTML file as the final output for a review session. Copies it to a new non-overwriting output path, preserves the source, and marks the session resolved. Use only after the agent has applied and visually verified remaining free-form annotations.
html_review_start_sessionwrite actionCreate an isolated local review session for an HTML file. Copies the source into session storage, starts a browser editor URL, and chooses a non-overwriting output path. Returns the session id and review URL.
Public scan report
scanner v0.1.10 · 2026-11-03 · same rubric, same numbers if you re-run it
- Code scan24 source files scanned25/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitylocal package, no credentials required12/15
- Maintenancelast push 73 days ago12/15
- Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
How Anno compares
Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 88/100
- higher than 75% of 48,255 servers
- Tools
- 6
- the typical server has 7
- Tools that change things
- 17%
- typical: 0%. Fewer write tools means less an agent can break
- GitHub stars
- 1
- more than 65% of 28,688 servers
- Downloads a week
- 57
- more than 47% of 12,191 servers with a package
What its tools let an agent do: 5 read · 1 change
Install Anno in Claude Code, Cursor or VS Code
Runs npx -y @philmingdao/anno on your machine. Read the scan report first; the gateway never runs local packages.
claude mcp add anno -- npx -y @philmingdao/anno
What the publisher says
From the Anno repository's README, as published. We do not edit it. Read it on GitHub
English · 简体中文 · 日本語 · 한국어 · Français · Español · Deutsch · Italiano · Português · ไทย
Anno
Anno is a local-first HTML review workspace for AI coding agents. It opens an isolated copy of a local HTML file in a browser, supports direct text and formatting edits, element comments, area annotations, and slide-aware review, then returns a durable handoff that an agent can claim and resolve into a verified standalone HTML file.
The repository contains one shared MCP server and host-neutral skill, native plugin manifests where a host supports them, and copy-ready MCP templates for Cursor, Google Antigravity, Windsurf, GitHub Copilot, and Meta Muse Code. DeepSeek Harness support is maintained separately in philmingdao/anno-dsh-native, an in-process native plug-in built with DeepSeek Harness; Muse Code support remains experimental.
Highlights
- Local-only HTTP editor bound to 127.0.0.1
- Source files are never overwritten
- Direct text, typography, color, position, page-note, element, and area editing
- Durable, idempotent agent handoffs
- Legacy compatibility with existing needs_codex sessions
- Shared MCP and SKILL.md implementation across supported hosts
- Simplified Chinese and English UI, light and dark themes
Screenshots
The review editor opens an isolated copy of your HTML in the browser. Edit text and styles directly, leave element or area comments, and hand the collected changes back to the agent in one click.
Requirements
- Node.js 22 or newer
- A host that supports local stdio MCP servers and can access local files
- A browser for the review editor
Supported agent tools
Codex, Claude Code, WorkBuddy, and CodeBuddy use packaged plugin manifests. Cursor, Google Antigravity, Windsurf, GitHub Copilot CLI/Chat, and Muse Code connect to the same local stdio MCP server through host-specific configuration templates. DeepSeek Harness uses the independent native anno-dsh-native repository, which integrates directly with the DSH profile, tool registry, and agent lifecycle without an MCP bridge.
See Agent tool integrations for copy-ready configuration and host limitations.
One-command install
No clone or build is required. The installer detects available agent tools, lets you choose them interactively, merges only the anno entry into existing JSON/JSONC configuration, installs the shared skill, creates a timestamped backup before changes, and runs an MCP handshake.
npx -y @philmingdao/anno@0.4.0 setupShortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
Anno: common questions
- Is Anno MCP server safe?
- Yes, by our scan: it is graded A (88/100). Read the Anno safety report
- How do I install Anno?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does Anno need an API key?
- Not as far as the registry entry and our scan can tell: no credentials are declared or required.
- Is Anno maintained?
- The last commit was 74 days ago (2026-08-22). The latest release is v0.4.0.
- How does Anno compare with other MCP servers?
- safety score 88/100 (higher than 75% of 48,255 servers); tools 6 (the typical server has 7); tools that change things 17% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.