backchat (openma-ai/backchat) is an MCP server listed on the M8ven Trust Index. It scores 68 out of 100, grade C. It declares 66 tools. No publisher has claimed this listing.

Caution

68/100

backchat

Item: backchat
Rating: 68
Author: M8ven

Local-first desktop workspace for ACP agents—DeepSeek Harness, Claude Code, Codex, Gemini CLI, OpenCode, Hermes, and OpenClaw—with projects, tools, files, browser, and sessions.

→ openma-ai/backchat · listed on github_repo_search

Caution. Specific findings reduced this grade. They are listed on the page. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.

How we verified

Code Verified⚡ Live Monitored: not connected

Verified is a snapshot. Live keeps it current, and builds your track record.

⚡ Connect GitHub → continuous verification on every push why connect →

Who stands behind it

openma-ai

Source: github_repo_search

Is this your MCP?

Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →

⚡ Connect GitHub for Live Monitored

Install from

GitHub

The grade above is for the source repository. Registries can serve a different version, so we mark the ones we were not able to read.

// key findings

🚨

Known vulnerabilities in dependencies: 1 critical, 3 high

Affects packages this MCP installs at runtime. Upgrade or remove the affected dependency.

// tools this server exposes66 tools

These names and descriptions are the publisher's own, read from the source code. We print them as written. Our assessment is the findings above, not this list.

browser_tabs

List, open, select, or close tabs in this task's visible Backchat browser window.

browser_navigate

Navigate the active tab in this task's visible Backchat browser window.

browser_screenshot

Capture the active in-app browser tab as PNG.

browser_click

Click a CSS selector, text= label, or :has-text() match in the active in-app tab.

browser_type

Type into an editable element in the active in-app tab.

browser_get_text

Read visible text from the active in-app tab or a matching element.

browser_eval

Evaluate JavaScript in the active in-app browser tab and return its result.

browser_close

Close the active tab in this task's in-app browser window.

browser.list

List available browser backends and their capabilities.

browser.documentation

Read Backchat Browser tool contract, safety rules, and parity notes.

browser.get

Read one browser backend descriptor by id, type, or alias.

browser.tabs

List tabs for a browser backend.

browser.selected_tab

Read the currently selected automation tab for a browser backend.

browser.user_open_tabs

List read-only top-level tabs already open in the user browser.

browser.get_tab

Read one browser tab by id without changing the selected automation tab.

browser.name_session

Name the current browser automation session for display and state snapshots.

browser.session_name

Read the current browser automation session name, if one was set.

browser.select_tab

Mark an existing browser tab as the selected automation tab.

browser.new_tab

Create a new tab in a browser backend.

browser.goto

Navigate an existing browser tab to an allowed URL.

browser.visibility_get

Read whether a browser backend is visible to the user.

browser.visibility_set

Show or hide a browser backend when it supports visibility.

browser.viewport_set

Resize a browser backend viewport when it supports viewport control.

browser.viewport_reset

Reset a browser backend viewport to its default size.

browser.reload

Reload a browser tab.

browser.back

Move a browser tab backward in history.

browser.forward

Move a browser tab forward in history.

browser.wait_for_url

Wait until a browser tab reaches an exact URL.

browser.wait_for_load_state

Wait until document.readyState reaches the requested load state.

browser.close_tab

Close a browser tab.

browser.title

Read the current title of a browser tab.

browser.url

Read the current URL of a browser tab.

browser.screenshot

Capture a screenshot of a browser tab.

browser.console_logs

Read console log entries observed for a browser tab.

browser.dom_snapshot

Read visible text from the page DOM.

browser.evaluate

Evaluate JavaScript in the page context.

browser.click

Click an element matched by a CSS selector.

browser.type

Set text on an input or editable element matched by a CSS selector.

browser.keypress

Send one keypress to the active page.

browser.cua_click

Click page coordinates using the browser CUA surface.

browser.dom_cua_snapshot

Read compact visible interactable DOM for DOM CUA.

browser.dom_cua_click

Click a DOM CUA node by node_id.

browser.locator_count

Count elements matched by a locator descriptor.

browser.locator_click

Click the first element matched by a locator descriptor.

browser.locator_fill

Set text on the first input or editable element matched by a locator descriptor.

browser.locator_press

Dispatch a keypress on the first element matched by a locator descriptor.

browser.locator_set_checked

Set checked state on the first checkbox or radio matched by a locator descriptor.

browser.locator_select_option

Select an option on the first select element matched by a locator descriptor.

browser.locator_inner_text

Read inner text from the first element matched by a locator descriptor.

browser.locator_attribute

Read an attribute from the first element matched by a locator descriptor.

browser.dialog

Read the currently active JavaScript dialog, if any.

browser.dialog_accept

Accept the currently active JavaScript dialog.

browser.dialog_dismiss

Dismiss the currently active JavaScript dialog.

browser.page_assets

Inventory page resource URLs discovered in a tab.

browser.bundle_assets

Download page assets into a Backchat-managed local bundle.

browser.clipboard_read_text

Read text from the host clipboard.

browser.clipboard_write_text

Write text to the host clipboard.

plugin_search_skills

Search Codex-compatible plugin workflows before starting work that may match an installed skill. Read a matching skill before following it.

plugin_read_skill

Read the complete SKILL.md instructions for a matching Codex-compatible plugin skill.

plugin_read_file

Read a relative reference, script, template, or asset text file from an installed plugin after its SKILL.md asks for it.

schedule_create

Create a one-time or recurring scheduled task using this task's harness and project.

schedule_list

List schedules created by this task.

schedule_update

Update, pause, or resume one of this task's schedules.

schedule_delete

Delete one of this task's schedules and its run history.

openma_sessions_list

List other local OpenMA sessions that can be referenced from the current task.

openma_sessions_read

Read the user and assistant conversation from another local OpenMA session by its stable session ID.

// known CVEs in dependencies1 critical3 high2 low

Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.

criticalelectron@42.3.1GHSA-q6m5-f73j-m9mc

Electron: Buffer performs incorrect byte length calculations resulting in heap buffer under/overflow

highnanoid@5.1.11GHSA-28wg-ghj8-5hjv

nanoid: non-secure generators can loop indefinitely with negative size

highsmol-toml@1.6.1GHSA-7w5x-hrqm-74c2

smol-toml: Denial of Service via malformed TOML documents

highsharp@0.35.2GHSA-rgj7-g3m4-5g8c

sharp: Vulnerabilities in libheif: GHSA-g89c-p67h-r497 and GHSA-2jg2-4ch7-h545

lowelectron@42.3.1GHSA-r4w5-6pfg-jxp5

Electron: ProtocolResponse.url reuses the default session cache instead of the registering session

Depend on this server? Get alerted when its CVEs change.Watch this server free →

// environment variables

To run this server yourself, you supply these values. They go in your own MCP client configuration and stay on your machine. The secret label means the value is sensitive, not that the server mishandles it.

configBACKCHAT_BROWSER_EXTENSION_PORT

configBACKCHAT_E2E_VISIBLE

configBACKCHAT_HOME

configBACKCHAT_TEST_HOOKS

configELECTRON_RENDERER_URL

configNPM_CONFIG_REGISTRY

configOMA_PROFILE

configOPENMA_ACP_BIN_DIR

configSHELL

// quality suggestions

Tool annotations

8/66 tools have annotations

Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.

All four hints declared on every tool

66/66 tools missing one or more hints — browser_tabs (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); browser_navigate (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); browser_screenshot (missing: destructiveHint, idempotentHint, openWorldHint), +63 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.

For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.

License file

No license file

Add a LICENSE file (MIT, Apache-2.0, etc.).

Shell command execution

1 call in production code run through a shell (src/main/session-cwd.ts:48)

Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.

Production dependencies are patched

0 critical, 2 high severity in production deps — nanoid@5.1.11 (high), smol-toml@1.6.1 (high)

Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.

Dev dependencies

2 critical/high in dev-only deps (does not ship to users)

Upgrade dev dependencies when convenient.

Dependency freshness

1/35 production deps stale: rrule@2023-11-10 (2.9y)

Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.

// full audit trail

The findings above are the summary. The full trail, every check we ran, each deduction, the network hosts observed and the dependency advisories, goes to verified publishers, along with an alert whenever a new one lands. Verified publishers can also review each finding and dispute it in one click. Publisher corrections have sharpened several of our checks this month, because the maintainer knows the codebase better than any scanner.

// improvement guidance — verified publishers only

We have 6 concrete improvements we can share with the publisher of this MCP. Each comes with specific guidance to raise the trust score.

// embed badge in your README

[![M8ven Verified](https://m8ven.ai/badge/mcp/openma-ai/backchat?variant=verified)](https://m8ven.ai/mcp/openma-ai/backchat)

Shows verification status without the grade. Want the grade badge instead? Remove ?variant=verified from the URL.

commit: 17d70628462be121cd1aadab0b5878fe2c89976b

code hash: 69289cf9133037f53cc1ae45c3df04fd3e269d144eb8ff109d9ccab7942db86b

view raw JSON →

Check MCPs from inside your assistant

Tool Check · MCP

Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.

https://m8ven.ai/api/mcp/tool-check

check_toolsearch_toolscompare_toolsrecommend_alternativescheck_publisherreport_concern

How to add it →Free · no account needed · works in any MCP client