Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Mcp (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Local stdio bridge to the Dock MCP server. Point any MCP-capable agent (Claude Desktop, Cursor, Windsurf, Zed, Cline, Continue) at your Dock workspaces in one config change.
Do you need this package?
>
If your agent supports remote MCP connectors (Claude.ai web, Claude.ai Projects), you don't need this — add https://trydock.ai/api/mcp as a custom connector and follow the OAuth flow. See the MCP reference.>
This package is for agents that only speak local stdio MCP — the dominant pattern for Claude Desktop and most code-editor agents today.
(trydock.ai).
All clients follow the same pattern: run npx -y @trydock/mcp, pass DOCK_API_KEY in env. Per-client JSON snippets are in configs/:
| Client | File | Typical config path |
|---|---|---|
| Claude Desktop | configs/claude-desktop.json | ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) |
| Cursor | configs/cursor.json | ~/.cursor/mcp.json |
| Windsurf | configs/windsurf.json | ~/.codeium/windsurf/mcp_config.json |
| Zed | configs/zed.json | ~/.config/zed/settings.json (under context_servers) |
| Cline (VS Code) | configs/cline.json | VS Code settings → cline.mcpServers |
| Continue | configs/continue.json | ~/.continue/config.json |
Example (Claude Desktop):
{
"mcpServers": {
"dock": {
"command": "npx",
"args": ["-y", "@trydock/mcp"],
"env": {
"DOCK_API_KEY": "dk_..."
}
}
}
}All 8 tools are forwarded to the hosted Dock server. JSON schemas live in schemas/.
| Tool | Purpose |
|---|---|
list_workspaces | Enumerate workspaces you can access |
get_workspace | Fetch a workspace by slug |
list_rows | Read rows from a table-mode workspace |
create_row | Append a row |
update_row | Partial-merge update |
delete_row | Remove a row |
create_workspace | Create a new workspace |
get_recent_events | Read the activity log |
Full reference with examples: trydock.ai/docs/mcp.
The bridge is a ~100-line Node process (src/bridge.js). Every JSON-RPC message your agent writes on stdin is forwarded over HTTPS to https://trydock.ai/api/mcp with your DOCK_API_KEY as Bearer auth. The hosted server owns authentication, rate limits, audit, and tool execution. The bridge stores no state and logs no request bodies.
Environment variables:
| Variable | Required? | Purpose |
|---|---|---|
DOCK_API_KEY | yes | Bearer token (get one from Settings → API keys) |
DOCK_MCP_URL | no | Override the upstream endpoint (staging / self-host). Default: https://trydock.ai/api/mcp |
each HTTPS call. It's never logged, never written to disk by this bridge.
401 immediately on revocation.
Dock DB leak wouldn't expose usable credentials.
threat model, the revocation runbook, and what data is audited.
MIT. Copyright © 2026 Vector Apps, Inc.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.