watchcheck
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@watchcheck Scan for Chinese enterprise monitoring tools"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
watchcheck 🔍
English | 简体中文
See what's actually running on your Mac — and who's watching.
watchcheck reads the processes on your Mac (read-only) and turns cryptic names into plain language: what each one is, who makes it, and — its specialty — whether it's endpoint-monitoring software (EDR / DLP / MDM / network & print auditing). It has first-class coverage of Chinese enterprise monitoring agents (深信服 Sangfor, 亿赛通 ESafeNet, IP-Guard, 奇安信, 360, 联软, 天空卫士, 北信源, and other domestic EDR/DLP/MDM tools) that Western tools — Little Snitch, KnockKnock, even general-purpose LLMs — consistently misidentify or don't know at all.
Two ways to use it
One read-only engine, two front-ends — pick either or both:
🖥️ Live panel | 🤖 MCP server | |
What | A local, auto-refreshing dashboard that reads your current processes — like Activity Monitor, but it explains each one and flags monitoring software | Plugs into your AI assistant (Claude, Cursor, …) so the LLM can read your live processes and answer questions about them |
For | Anyone — no AI, no account, no setup beyond install | People who live in an AI client and want to ask in their own words |
Run |
| add to your MCP config, then ask Claude |
Network | none — binds | none — local stdio |
watchcheck is read-only and honest by design. It identifies software and describes what that class of software is capable of per vendor docs. It does not prove any tool is actively capturing you right now, and it cannot see the content of any data being sent. It is a transparency tool, not a way to evade legitimate corporate policy. On a company-managed device, removing or tampering with required software may violate your employment agreement.
Related MCP server: macos-control-mcp
Install
Requires Python 3.10+ and macOS.
# with uv (recommended)
uv tool install watchcheck # once published
# or from source
git clone https://github.com/derkcc/watchcheck && cd watchcheck
uv venv --python 3.12 && uv pip install -e .🖥️ The live panel
A local, read-only dashboard that re-collects your processes / CPU / memory / GPU
every couple of seconds and explains them. Binds 127.0.0.1 only — never touches
the network, never modifies anything.
watchcheck panel # opens http://127.0.0.1:8787/
watchcheck panel --lang en --interval 2 --port 8787Activity-Monitor-style tabs — Monitoring / CPU / Memory / GPU / All processes — where every process row carries an inline plain-language explanation and a 🟢/🔴/⚪ marker; monitoring software is flagged with its capabilities and evidence.
Prefer a static, shareable file instead of a live server?
watchcheck report # one-shot HTML snapshot → ~/watchcheck-report.html
watchcheck report --lang en # English (~/watchcheck-report.en.html)Both are bilingual (--lang zh|en). GPU is reported system-wide — macOS exposes
no per-process GPU without sudo.
🤖 The MCP server
Let your AI assistant read and explain your live processes. Add to your MCP client
config — Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"watchcheck": { "command": "watchcheck" }
}
}From source (no install):
{
"mcpServers": {
"watchcheck": {
"command": "uv",
"args": ["--directory", "/path/to/watchcheck", "run", "watchcheck"]
}
}
}Then just ask:
"Scan my Mac — is my company monitoring me, and what can they see?" "What is
acnvmagent?" "What monitoring tools does watchcheck know about?"
How it works (no screenshots needed)
You never copy process names or paste screenshots. The server runs on your Mac and reads the live process list itself; Claude calls it and explains the result.
flowchart TD
A["You — ask in plain language<br/>(no screenshots, no copy-paste)"] --> B["Claude picks a tool:<br/>scan / overview / explain_process"]
B --> C["watchcheck runs locally on your Mac<br/>reads processes via ps / launchd / certs<br/>read-only · no network · nothing modified"]
C --> D["Returns structured facts:<br/>vendor / type / capabilities / CPU · memory<br/>(things it doesn't know are marked 'unknown')"]
D --> E["Claude explains in plain language<br/>and answers follow-ups"]
E --> ADivision of labor: watchcheck reads the processes and supplies the facts
(from its signature DB); Claude orchestrates the calls, turns the facts into
plain language, and fills in anything marked unknown from its own knowledge.
Tools exposed
Tool | What it does |
| Read-only scan → identified monitoring software with evidence, capabilities, privacy impact |
| Typed breakdown of everything running (Apple system / browser / cloud / your own VPN / monitoring / unknown …), duplicates collapsed, with CPU/memory/GPU |
| Explain one process / label / bundle id in plain language |
| The full catalog of what watchcheck can identify (transparency) |
| Raw collected artifacts, no matching (for investigating unknowns / contributing) |
How it works
watchcheck reads only what macOS already exposes — nothing is modified, no files are read for content, no network calls:
Source | Command | What it reveals |
Processes |
| Running agents + CPU / memory |
Persistence | LaunchDaemons/Agents plists | What auto-starts |
System extensions |
| Network / endpoint-security filters |
Kernel extensions |
| Kernel-level agents (highest privilege) |
MDM |
| DEP / MDM enrollment |
Certificates |
| Corporate root CAs (HTTPS interception) |
GPU |
| System-wide GPU utilization |
It then matches these against two data files: a curated, community-maintained
monitoring signature DB
(signatures.yaml — the part that knows
Chinese enterprise tools) and a common-process catalog
(common_processes.yaml — everyday
macOS processes), so it can reassure you that most of what's running is normal
and clearly flag what isn't. The signature DB is the whole point; everything else
is a thin, replaceable shell.
Contributing signatures (the important part)
Coverage of Chinese enterprise tools on macOS is the gap, and it's where you can
help most. If raw_inventory (or the panel's "unknown" rows) shows something
watchcheck doesn't recognize:
Find the artifact (process name, launchd label, bundle id, kext id, cert CN, path).
Add an entry to
signatures.yamlfollowing the schema and the honesty rules at the top of that file.Set
verified: trueonly if you confirmed it on a real machine.Open a PR. See CONTRIBUTING.md.
Signatures are facts about software, contributed by people who see it in the wild. That's the moat — and it only grows with help.
Roadmap
Windows + Linux collectors
Optional
outbound_activity(which monitoring processes have live connections — volume/destination only, never content)Wider Chinese-vendor macOS signatures
Per-process CPU sparklines in the live panel
License
MIT. See LICENSE.
Vendor and product names are used nominatively to identify software. No affiliation with or endorsement by any vendor is implied.
Available Tools
5 toolsexplain_processA
Explain one process / launchd label / bundle id in plain language.
Built for non-technical users who paste a single name from Activity Monitor
(e.g. "WindowServer", "EasyConnect", "GoogleSoftwareUpdate"). Resolution order:
monitoring signature → common benign catalog → returns unknown so the host
LLM can explain the long tail. Always answers something useful.
Inputs: name (required). Returns: object
list_signaturesA
List everything watchcheck can currently identify (transparency).
Returns the signature catalog (vendor, product, category, origin, coverage) without the raw match tokens. Use this to tell the user what the tool knows — and what gaps remain (especially Chinese enterprise tools).
Inputs: no named parameters. Returns: object
overviewA
Friendly typed overview of EVERYTHING running on this Mac (read-only).
For users who don't understand processes: classifies every running process into types (Apple system / browser / cloud-sync / updater / communication / your own VPN / monitoring / unknown ...), collapses duplicates (e.g. 30 Chrome helpers -> 1), and flags monitoring software. Present the type counts first, reassure that most are normal, then call out anything flagged.
Inputs: no named parameters. Returns: object
raw_inventoryA
Return the raw read-only inventory with NO matching (power users).
Dumps processes, launchd jobs, extensions, kexts, certs and MDM status as collected. Use this to investigate something watchcheck didn't recognize, or to gather artifacts for a new signature contribution.
Inputs: no named parameters. Returns: object
scanA
Scan this Mac (read-only) and report identified monitoring software.
Inspects running processes, launchd jobs, system/kernel extensions, MDM enrollment and trusted certificates, then matches them against the watchcheck signature database. Returns identified EDR/DLP/MDM/network-audit software with evidence, capabilities and privacy impact. Group the result by category and explain each finding in plain language for the user.
Inputs: no named parameters. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 5 tools
Each tool has a distinct purpose: explain_process explains a single process, list_signatures shows the catalog, overview gives a high-level summary, raw_inventory provides raw data for power users, and scan performs a full monitoring detection. No overlap.
Tool names use snake_case but mix conventions: verb_noun (explain_process, list_signatures), single word (overview, scan), and noun phrase (raw_inventory). While readable, the pattern is inconsistent.
Five tools is well-scoped for the server's purpose of monitoring software detection. Each tool serves a clear role without redundancy, fitting within the ideal 3-15 range.
The tool set covers the core workflow of scanning, explaining, and listing known signatures. A minor gap exists in advanced features like diffing scans, but the domain is well-covered for transparency and detection.
Maintenance
Related MCP Connectors
Connect ChatGPT and Claude to your Mac, Windows or Linux with secure MCP remote access.
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
- DNS-DATA RemoteDeprecatedcom.dns-data
Connect AI assistants to authorized Windows and macOS devices for files, commands, and management.
Remote MCP server exposing your Node Command fleet (files, shell, processes, browser) to AI hosts.
Related MCP Servers
- AlicenseAqualityDmaintenanceAn MCP server that lets AI assistants see your macOS desktop, capture screenshots, read browser tabs, and preview files via local macOS tools.48 npmMIT
- AlicenseBqualityDmaintenanceMCP server that enables AI to fully control macOS — mouse, keyboard, terminal, screenshots, window management, UI element detection, and provides AI-optimized information reporting.3621 npmMIT
- FlicenseNot gradedqualityDmaintenanceEnterprise-grade macOS MCP server that gives AI agents eyes, hands, and a terminal, enabling screen capture, mouse/keyboard control, shell commands, and file operations.-
- AlicenseBqualityBmaintenanceEnables AI agents to directly access native macOS services, media, system health, and administration tools through a local MCP server.4235 npm2MIT