dsh-mcp-host

Expose any stdio MCP server as native DeepSeek Harness tools. Host-only plugin, no browser UI, no npm dependencies.

Built for one job: let a DSH agent drive a browser through the official Microsoft Playwright MCP server, attached to the Edge/Chrome you are already using.

0.4.0 — the browser tools unload when the browsing turn ends (and right after edge_close); later requests carry only edge_tools. The MCP process and the open browser session stay alive, so reloading is one cheap call. See Unload after use.

What it does

1. On load, spawns the MCP server once (default npx --yes @playwright/mcp@latest --cdp-endpoint=msedge) and calls initialize + tools/list.
2. Registers one bootstrap tool, edge_tools (name configurable). Its description is deliberately short and states when to call it — and when not to.
3. When the model calls edge_tools, the bridge registers the real edge_* tools for that Agent only, hides the bootstrap entry, and forwards every call to MCP tools/call. Screenshots are written to a temp file and returned as a path, not as base64.
4. Unload after use: when that turn ends (session/event → turn/end), the per-Agent tools are removed and edge_tools becomes visible again, so every later request is back to one small schema. edge_close releases immediately, and an idle countdown (MCP_BRIDGE_RELEASE_MS, default 3 min) is the backstop. The MCP process and the open browser session stay alive, so reloading is one cheap call. On agent/disposed (and on plugin dispose) the same teardown runs.

Tools are renamed: browser_navigate becomes edge_navigate (prefix configurable).

Lazy loading (default)

The 12-tool default set is not in the prompt until a browsing task needs it. A session that never touches a browser pays for one small schema (edge_tools), not twelve.

• Set MCP_BRIDGE_EAGER=1 to restore the old behaviour (register every tool at boot).
• Activation is per Agent: the parent session and each subagent must call edge_tools independently.
• After activation the bootstrap entry is hidden for that Agent (tools.restrict({ deny })), so the prompt goes back to a single, smaller surface.

Unload after use (default)

Loading is temporary, so a session that browsed once does not carry twelve schemas forever:

• Primary trigger: the end of the turn. As soon as the session emits turn/end, that Agent's tools are disposed and edge_tools appears again — the next request only pays for the small bootstrap schema. The next browsing turn reloads them with one call.
• edge_close: immediate unload (the page is closed, so the burst is over).
• Idle backstop: if no turn/end ever arrives (non-interactive hosts, long-lived agents), the tools also unload after MCP_BRIDGE_RELEASE_MS without a browser call. 0 disables the backstop.
• The MCP server process is not killed and the browser stays open — only the prompt surface shrinks, and reloading keeps tabs, logins and history.
• MCP_BRIDGE_RELEASE_ON_TURN_END=0 (with MCP_BRIDGE_RELEASE_MS=0) restores the 0.2.0 behaviour: tools stay for the whole Agent lifetime.

Default tool set (12 of Playwright's 25)

edge_navigate, edge_navigate_back, edge_snapshot, edge_click, edge_type, edge_fill_form, edge_press_key, edge_tabs, edge_take_screenshot, edge_evaluate, edge_wait_for, edge_close.

Set MCP_BRIDGE_ALL=1 to expose everything the server offers.

Screenshots: image, path, or both

Playwright MCP ≥ 0.0.83 saves every screenshot to disk and returns text containing the path ([Screenshot of viewport](.playwright-mcp\page-....png)), plus an image part unless told otherwise. The path is relative to the MCP server's working directory (on this machine the profile dir) unless MCP_BRIDGE_COMMAND sets an absolute --output-dir.

This plugin never forwards that image part. normalizeResult writes images to a temp file and every tool renders JSON text, so the model sees a path (~26 tokens), not an image. A screenshot has always been cheap here.

So MCP_BRIDGE_IMAGES does not save context in DSH. What it buys:

• MCP_BRIDGE_IMAGES=omit — the upstream stops sending the image at all: ~12 K fewer base64 characters over stdio per screenshot, and one less duplicate temp file. Pick it when the agent never looks at the picture. Caveat: with omit the only path is the relative one inside the text, because the plugin's absolute temp copy (images[]) only exists in the default mode — an agent that wants to open screenshots should keep allow.
• MCP_BRIDGE_IMAGES=only — keep the image, drop the accompanying text.
• Appended only to Playwright MCP command lines (@playwright/mcp or playwright-mcp); other servers are left untouched, and an explicit --image-responses inside MCP_BRIDGE_COMMAND always wins.
• Default is unchanged (allow).

Context only suffers where a client renders MCP image content (Claude Code, Cursor, …): there a 1280×720 capture is ~1.2k tokens. Those clients do not use this plugin — set --image-responses=omit in their own MCP config instead.

Output hygiene

Playwright MCP writes every screenshot and accessibility snapshot into its output directory (--output-dir, default <cwd>/.playwright-mcp) and never rotates it, so a long browsing life leaves hundreds of files behind. The bridge prunes once at plugin load and after every tool call: it keeps the newest MCP_BRIDGE_KEEP_OUTPUT files in that directory and in its own temp image directory (%TEMP%\dsh-mcp-bridge), deleting older .png/.jpg/.jpeg/.webp/.yml/.yaml/.log files. Anything else in those directories is left alone, and the screenshot just taken is always the newest, so reading it still works. 0 deletes all managed files; a missing directory or an unparsable value is a no-op.

Install

From npm (published as dsh-mcp-host):

dsh plugin --profile web add dsh-mcp-host

From a local checkout (development):

cd E:\dsh
.\rollback.cmd guard add "file:E:\dsh\dsh-mcp-bridge"
.\rollback.cmd health

Restart DSH (or rely on hot-load) and open a new session. The default MCP command uses npx, so Node.js (which ships npx) must be on PATH; the first call downloads @playwright/mcp from npm.

Configuration

Env var	Default	Meaning
MCP_BRIDGE_COMMAND	npx --yes @playwright/mcp@latest --cdp-endpoint=msedge	The MCP server command line (any stdio MCP server)
MCP_BRIDGE_PREFIX	edge	Tool name prefix
MCP_BRIDGE_BOOTSTRAP	edge_tools	Name of the single lazy-loading entry tool
MCP_BRIDGE_RELEASE_ON_TURN_END	1	0 = do not unload at turn end (idle backstop only)
MCP_BRIDGE_RELEASE_MS	180000	Idle ms before the browser tools unload themselves again (backstop; 0 = off)
MCP_BRIDGE_EAGER	unset	1 = register every tool at boot (old behaviour, no bootstrap)
MCP_BRIDGE_ALL	unset	1 = expose every tool the server offers
MCP_BRIDGE_IMAGES	unset	omit = upstream sends no image (the text keeps the path; saves the transfer and a duplicate temp file — no context change in DSH), only = drop the text; Playwright MCP commands only
MCP_BRIDGE_KEEP_OUTPUT	20	Newest N files kept in Playwright's output dir after each call; 0 = delete all managed files
MCP_BRIDGE_TOOLS	unset	Comma-separated allow-list of MCP tool names
MCP_BRIDGE_TIMEOUT	120000	Per-call timeout (ms)
MCP_BRIDGE_INIT_TIMEOUT	30000	Startup timeout (ms)

Prerequisite for the Edge/Chrome you are already using

Open edge://inspect/#remote-debugging (or chrome://inspect/#remote-debugging) and enable Allow remote debugging for this browser instance. Then --cdp-endpoint=msedge (or =chrome) attaches to that running browser with your logins and extensions. Without the toggle, Playwright launches a fresh browser instead.

Failure behaviour

If the MCP server cannot start or does not answer, the plugin registers nothing and logs the reason — it never throws, so it cannot take the host boot down. Check MCP_BRIDGE_INIT_TIMEOUT and the stderr tail printed in the DSH console.

Privacy

The bridge itself makes no network calls. It spawns exactly one child process (the MCP server you configured) and forwards bytes between it and the tool call. With --cdp-endpoint, the MCP server attaches to your existing browser, which means it can see and act on the tabs you have open — treat that as giving the agent your browser session.

License

MIT.