kitty-hive (seangx/kitty-hive) is an MCP server listed on the M8ven Trust Index. It scores 69 out of 100, grade C. It declares 35 tools. No publisher has claimed this listing.

Caution

69/100

kitty-hive

Item: kitty-hive
Rating: 69
Author: M8ven

MCP server for multi-agent collaboration enabling AI agents to communicate, delegate tasks, and share artifacts across clients and machines with federation support.

→ seangx/kitty-hive · listed on glama

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

⚡ 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

seangx

Source: Glama

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

Glamanot scored here

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

⚠️

Tool descriptions don’t match what handlers do

1 tool describes read intent but its handler mutates — hive_file_fetch (line 119: mkdirSync(dirname(dest), { recursive: true }))

✅

No credential exfiltration, no sensitive file access, no obfuscation

Static analysis found nothing flowing your secrets to unexpected places.

🔐

You'll be asked for 1 credential: HIVE_AGENT_KEY

These are read from process.env at runtime. Make sure you trust where they’ll be sent.

// tools this server exposes35 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.

hive_start

hive_whoami

Show the agent currently bound to this session (or via `as`).

hive_update_role

hive_rename

Change your global display_name (display only — addressing uses id).

hive_codex_pane_ws

hive_opencode_pane_server

hive_agents

List all known agents on this hive. Use this to find agent ids for cross-team DM/task.

hive_dm

hive_dm_read

hive_file_fetch

Fetch an attachment by file_id (from a DM you received). Returns a `path` on the hive node serving you (your local hive when running locally; the receiver's hive in federated setups — already replicated). Pass `save_to` to copy it to a path of your choice (a trailing "/" treats it as a directory and…

hive_inbox

Check unread DMs, team events, and task events. Marks returned items as read.

hive_peers

List all known federation peers (each entry includes name, status active/inactive, exposed agents, last_seen).

hive_remote_agents

List agents that a remote peer has exposed to us. Cached 5 minutes; pass fresh=true to bypass. After getting an agent id from here, address them in DM/task as "<agent_id>@<peer-name>".

hive_task

hive_task_claim

Claim an unassigned task (status "created", no assignee).

hive_task_cancel

Cancel a task. Creator-only. Works in any non-terminal state (created, proposing, approved, in_progress, awaiting_approval). Notifies the assignee so they stop work.

hive_task_complete

hive_tasks

List tasks. Without `team`: tasks you created or are assigned to. With `team`: all tasks bound to that team via source_team_id (you must be a current team member; non-members get a 403). Use this before creating a task to avoid duplicating in-flight work in the team.

hive_check

Check the current state of a task by id. For a bound agent, fetching the full task also advances that task's read cursor through the returned current seq.

hive_workflow_propose

hive_workflow_approve

Approve a proposed workflow (creator only). Starts step 1.

hive_workflow_step_complete

Mark your part of the current step as complete. If the step has `gate: true` and you were the last assignee, the task enters `awaiting_approval` instead of auto-advancing — the creator must call hive-workflow-step-approve to release the gate.

hive_workflow_step_approve

hive_workflow_reject

Reject the current step. Sends task back to a previous step.

hive_team_create

Create a new team. Optionally set your nickname in this team.

hive_team_join

Join an existing team by name or id.

hive_team_list

List all teams on this hive (with their members).

hive_team_info

Get detailed info about a team you are in (members with nicknames + recent events).

hive_team_events

Fetch events from a team. Use "since" for incremental polling.

hive_team_message

Post a message to a team (broadcasts to all members).

hive_team_set_rules

Replace a team's rules / charter (free-form markdown shown to all members on hive_start + hive_team_info). HOST-ONLY: only the team's host agent can call this; non-hosts are rejected. Pass empty string to clear. The change is pushed to all members as a `team-rules-update` notification — they should …

hive_teams

List teams you are a member of (with your nickname in each).

hive_team_rename_nickname

hive_team_leave

hive-whoami

Show your agent id, display_name, and registration info. If not registered, pass `name` to register.

// 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.

configCODEX_APPSERVER_CWD

configCODEX_CHANNEL_MODE

configCODEX_CMD

configCODEX_EXTRA_ARGS

configCODEX_PROFILE

configEDITOR

configHIVE_AGENT_IDSet HIVE_AGENT_NAME=<name> (or =<id>) in the env to skip this and auto-register on launch.

🔐 secretHIVE_AGENT_KEYAuto-register via opaque external_key (orchestrator-assigned; idempotent — same key returns same agent)

configHIVE_AGENT_NAMESet =<name> (or HIVE_AGENT_ID=<id>) in the env to skip this and auto-register on launch.

configHIVE_AGENT_ROLES

configHIVE_AGENT_SESSION_ID

configHIVE_AGENT_THREAD_ID

configHIVE_EVENT_MODECodex event ownership: auto starts daemon turns; foreground persists notifications without background inference

configHIVE_SUPERVISOR_PID

configHIVE_URLhive HTTP endpoint (default http://localhost:4123/mcp)

configKITTY_WAKEUP_SOCKET

configOPENCODE_CMD

configOPENCODE_SERVER_CWD

configSystemRoot

configVISUAL

// quality suggestions

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

35/35 tools missing one or more hints — hive_start (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); hive_whoami (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); hive_update_role (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +32 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.

Descriptions match behaviour

1 tool describes read intent but its handler mutates — hive_file_fetch (line 119: mkdirSync(dirname(dest), { recursive: true }))

Rename the tool, rewrite the description, or move the side-effect into a separate clearly-named tool.

Tool inputs are validated

31/34 tool handlers declare input schemas (91%)

Declare an inputSchema with zod/joi/yup on every tool definition.

License file

No license file

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

Tool test coverage

Only 0/35 tools referenced in tests (0%)

Write tests that reference each tool by name so every tool has at least one test.

Shell command execution

9 calls in production code run through a shell (src/index.ts:295, src/index.ts:348, src/index.ts:360)

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

Tool description accuracy

hive_file_fetch: 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.

// 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 8 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/seangx/kitty-hive?variant=verified)](https://m8ven.ai/mcp/seangx/kitty-hive)

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

commit: 9c74d88794d0addd1fcdded1f497b811f7bc79f9

code hash: b3d9d7fc3b19faead3de5ab0f197b9edf219f29df1ec4b9d65380a09820b7164

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