mcp-agent-bridge
A local MCP server that provides shared, real-time context across multiple AI agents via WebSocket and MCP resource notifications, enabling collaborative workspaces, memory, tasks, and messaging.
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
Verified is a snapshot. Live keeps it current, and builds your track record.
⚡ Connect GitHub → continuous verification on every pushwhy connect →Who stands behind it
xfajarr
Source: Glama
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 →
Install from
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.
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.
create_contextCreate a new named context (workspace). All agents that join it share the same memory, tasks, messages, etc.
list_contextsList all available contexts (workspaces) and who is currently in each one
join_contextSwitch your active context. All subsequent operations (memory, tasks, messages, etc.) will be scoped to this context.
leave_contextLeave current context and return to 'default'. Your data in the context is preserved.
delete_contextDelete a context and all its data. Cannot delete 'default'.
context_statusShow your current context, peers, and a summary of what's in it
identifyRegister this agent session. Call this first. Then use join_context() to enter a shared workspace.
get_eventsGet events from current context since a timestamp. For real-time push subscribe to ws://localhost:3721/ws.
send_messageSend a message to another agent in the current context
read_messagesRead your pending messages in the current context
broadcastSend a message to all agents in the current context
create_taskCreate and assign a task within the current context
list_tasksList tasks in the current context
update_taskUpdate a task in the current context
set_memoryWrite a value to the current context's shared memory. All peers get notified in real-time.
get_memoryRead from the current context's shared memory. Omit key to list all.
delete_memoryDelete a key from the current context's shared memory
save_artifactSave a code or text artifact in the current context
get_artifactGet a shared artifact by name from the current context
list_artifactsList all artifacts in the current context
add_to_discussionAdd a message to a shared discussion thread in the current context
get_discussionGet discussion threads in the current context
start_sessionStart a collaborative session in the current context. All peers are notified in real-time.
get_sessionGet the active session in the current context
end_sessionEnd the active session in the current context
agent_statusSee all connected agents, grouped by context
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
ws: Memory exhaustion DoS from tiny fragments and data chunks
ws: Uninitialized memory disclosure
Tool annotations
No tools have read-only/destructive annotations
Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.
All four hints declared on every tool
26/26 tools missing one or more hints — create_context (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); list_contexts (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); join_context (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +23 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.
Destructive tools are labelled
1 tool perform destructive updates without destructiveHint — delete_context deletes at line 365 (fs.unlink(dataFile(`ctx_${name}`)))
Add destructiveHint:true to any tool whose handler calls .delete(), .upsert(), .update(), unlink, rm, DELETE, DROP, REPLACE INTO, or any operation that overwrites existing data.
Descriptions match behaviour
1 tool describes read intent but its handler mutates — read_messages (line 52: fs.writeFile(dataFile(name), JSON.stringify(data, null, 2), "utf-8"))
Rename the tool, rewrite the description, or move the side-effect into a separate clearly-named tool.
License file
No license file
Add a LICENSE file (MIT, Apache-2.0, etc.).
Tests exist
No test files found
Add tests that exercise each declared tool.
Production dependencies are patched
0 critical, 1 high severity in production deps — ws@8.20.0 (high), ws@8.20.0 (low)
Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.
Tool description accuracy
read_messages: description implies read-only but handler writes/deletes/executes
Update tool descriptions to accurately reflect all capabilities — especially write, delete, or execute operations.
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.
[](https://m8ven.ai/mcp/xfajarr/shared-context-mcp)?variant=verified from the URL.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