MarkViewSAFE
Native macOS markdown preview + MCP server for Claude Code. Swift/SwiftUI, GFM, Mermaid, syntax highlighting. No Electron.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://github.com/paulhkang94/markview/releases/latest) &color=green) [](https://glama.ai/mcp/servers/@paulhkang94/markview)
Native macOS markdown preview with MCP server for Claude Code. Claude writes markdown — MarkView renders it live, in a real native window, while you work.
Versions: The macOS app (appbadge) and the npm MCP wrapper (npmbadge) are versioned independently. App releases happen when the Swift binary changes; npm patches happen for MCP server improvements. Both badges always show the latest of each.
Quick Start — Claude Code
One command to wire MarkView into every Claude Code session:
claude mcp add --transport stdio --scope user markview -- npx mcp-server-markview
That's it. Claude can now call preview_markdown to render any markdown string in a native macOS window, or open_file to open any .md file directly.
Claude Desktop Setup
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"markview": {
"command": "npx",
"args": ["mcp-server-markview"]
}
}
}Note: MCP servers belong in~/.claude.json(Claude Code) orclaude_desktop_config.json(Claude Desktop), not~/.claude/settings.json.
Installation
Homebrew (recommended)
``
f60e16ad7a37OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add mcp-server-markview -- npx -y [email protected]
Exposed tools (9)
8 read · 1 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
get_changed_files | read | List all changed files in a git repository (staged, unstaged, and untracked). |
get_word_count | read | Count words, characters, lines, and estimated tokens in markdown content. |
lint_content | read | Lint raw markdown content using MarkView |
lint_file | read | Lint a markdown file using MarkView |
open_file | read | Open an existing markdown file from disk in MarkView. |
outline | read | Extract the heading tree from markdown content. Returns a JSON array of headings |
preview_markdown | read | Render a markdown string in MarkView |
render_diff_file | write | Run git diff on a repository and render the output in MarkView with diff2html |
render_diff_raw | read | Render a raw unified diff string in MarkView with diff2html syntax highlighting. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (7 observation(s))
- Network
- declared (9 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (13)
AppIcon.icns
.public-docs.json
cd Tests/playwright && npx playwright test && printf 'TIER=test\nTS=%s\n' "$$(date +%s)" > ../../.last-verify-at && cd ../.. && bash scripts/bundle.sh --install
self.plan["refs"] = ["refs/tags/v../../bad"]
'arguments': {'content': 'test', 'filename': '../../etc/passwd'}@playwright/test
@modelcontextprotocol/sdk
@cloudflare/workers-types, typescript, wrangler
Sources/MarkViewCore/Resources/mermaid.min.js
Tests/playwright/__snapshots__/chromium-dark/golden-corpus.spec.ts/golden-corpus-light.png
Tests/playwright/__snapshots__/chromium/golden-corpus.spec.ts/golden-corpus-light.png
Tests/playwright/fixtures/alerts.html
Tests/playwright/fixtures/code-blocks.html
Gates applied: no_behavioural_pass.
f60e16ad7a37full audit observations/trust-audit/mcp-server/paulhkang94__markview.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | f60e16ad7a37 | SAFE | B | 89 | first audit |
Questions
What is the MarkView MCP server?
Native macOS markdown preview + MCP server for Claude Code. Swift/SwiftUI, GFM, Mermaid, syntax highlighting. No Electron.
What tools does MarkView expose?
9 in total: 8 read-only, 1 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is MarkView safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does MarkView need?
It reads GITHUB_TOKEN from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does MarkView run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as markview-mcp-worker at 1.0.0.
How current is this page?
The grade is for one exact copy of the source (f60e16ad7a37), read on 2026-10-08. The repository is watched and re-audited when it changes.