claude-code-studio
A fully functional web workspace for Claude Code CLI—chat, Kanban task board, task scheduling, multi-agent orchestration, MCP servers, skills, remote access (Web, SSH, Telegram), projects, and real-time data streaming.
Warning. Serious findings were identified. Review the full report before connecting. 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
Lexus2016
Source: github_repo_search
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.
ask_userAsk the user a question and wait for their response. Use this when you need clarification, a choice, or any input from the user before proceeding. Execution pauses until the user responds. Supports multiple questions (paginated), rich options with descriptions, and an "Other" free-text fallback on e…
message_botHand a task to another bot from your roster. The bot runs AFTER you finish your turn, so you will NOT see its answer inside this turn — do not wait for it and do not build your reply around its output. Write "task" so it stands on its own: the callee does not share your context and sees nothing exce…
notify_userSend a non-blocking notification to the user. Use this to report progress, milestones, warnings, or errors WITHOUT pausing execution. You continue working immediately after calling this tool. Use sparingly — only for meaningful status changes, not every minor step.
set_ui_stateUpdate the UI state (mode, model, agent). Use this when you transition between phases (e.g., from planning to execution) so the UI reflects your current state. This is fire-and-forget — execution continues immediately.
create_taskCreate a new task in the system. By default the task is queued for execution; pass status="backlog" (or "done" for an already-completed item) to add it to the Kanban board WITHOUT running it — use that when importing an existing plan, roadmap, tasks/ folder or checklist into the board. The workdir i…
create_chainCreate a chain of sequential tasks in one call. Tasks run in order by default (each depends on the previous). All tasks share a session for context continuity. Use depends_on_index to create custom dependency graphs within the chain.
list_tasksList tasks in your project. Useful for checking existing tasks, avoiding duplicates, or monitoring progress of child tasks you created.
get_current_taskGet details about your current task, including the context passed by the parent task. Call this first to understand your mission and any context provided.
report_resultStore a structured result for this task. Other tasks that depend on you can read this via get_task_result. Use this to pass data to downstream tasks — issue lists, PR URLs, analysis results, etc.
get_task_resultRead the structured result of a completed task. Use this to get output from tasks you depend on or from child tasks you created.
cancel_taskCancel a pending or running task. Use when you determine a task is no longer needed (e.g., the issue was already fixed, or a sibling task made it redundant).
check_user_messagesCheck if the user sent any clarifications or corrections while you are working. Returns pending messages or empty if none. Call this between major steps to stay aligned with user intent.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
multer vulnerable to Denial of Service via oversized array index in field names
multer vulnerable to Denial of Service via file descriptor leak on aborted uploads
multer vulnerable to Denial of Service via crafted multipart field names
multer vulnerable to file size limit bypass via async fileFilter race condition
APP_DIRASK_USER_SECRETASK_USER_SERVER_URLASK_USER_SESSION_IDBOTS_CALLERBOTS_SECRETBOTS_SERVER_URLBOTS_SESSION_IDBOTS_TURNCCS_ALLOWED_ORIGINSCCS_CONFIG_PATHconfig.json and .env on the data volume ( / CCS_ENV_PATH)CCS_ENV_PATHconfig.json and .env on the data volume (CCS_CONFIG_PATH / )CCS_GIT_CLONE_TIMEOUT_MSCCS_INTERRUPT_FILE_TTL_MSCCS_INTERRUPT_SECRETCCS_INTERRUPT_SESSIONCCS_INTERRUPT_URLCCS_MAX_CLARIFY_TURNSCCS_NODE_CMDCCS_REMOTE_EXEC_HOOKCCS_REMOTE_EXEC_MAX_BYTESCCS_REMOTE_EXEC_TIMEOUT_MSCCS_REMOTE_FILES_MAX_BYTESCCS_REMOTE_FILES_MAX_ENTRIESCCS_REMOTE_IMPORT_MAX_BYTESCCS_REMOTE_IMPORT_MAX_TOTALCCS_REMOTE_INITCCS_SSH_HOST_KEY_POLICYCCS_TASK_MANAGER_SECRETCCS_TMUX_FRAME_MSCLAUDE_HARD_CAP_MSCLAUDE_IDLE_TIMEOUT_MSCLAUDE_PROMPT_GRACE_MSCLAUDE_STARTUP_PROMPT_WAIT_MSCLAUDE_TIMEOUT_MSCLEANUP_INTERVAL_HOURSHOSTyour phone or another machine, set =0.0.0.0 — the console then warns you on everyINTERRUPT_SECRETINTERRUPT_SERVER_URLINTERRUPT_SESSION_IDLOG_LEVELMAX_TASK_WORKERSMULTI_AGENT_CONCURRENCYMULTI_AGENT_MAX_PLANMULTI_AGENT_MAX_TURNS_CAPNOTIFY_SECRETNOTIFY_SERVER_URLNOTIFY_SESSION_IDSESSION_SECRETSESSION_TTL_DAYSSET_UI_STATE_SECRETSET_UI_STATE_SERVER_URLSET_UI_STATE_SESSION_IDSSH_AUTH_SOCKTASK_DISCONNECT_TIMEOUT_MSTASK_MANAGER_SECRETTASK_MANAGER_SERVER_URLTASK_MANAGER_SESSION_IDTASK_MANAGER_TASK_IDTRUST_PROXYWORKDIRPORTTool 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
12/12 tools missing one or more hints — ask_user (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); message_bot (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); notify_user (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +9 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.).
No eval / new Function
1 eval() or new Function() call — dynamic code execution
Replace eval / Function with explicit parsing or safer alternatives.
Shell command execution
11 calls in production code run through a shell (server.js:441, server.js:8748, server.js:8756)
Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.
Readable source code
1 file are minified or bundled, which is usually build output rather than concealment
Ship unminified, readable source.
No arbitrary install scripts
Has postinstall/preinstall script — runs arbitrary code on npm install
Remove postinstall/preinstall hooks unless they’re essential.
Production dependencies are patched
0 critical, 3 high severity in production deps — multer@2.2.0 (high), multer@2.2.0 (high)
Run npm audit fix, or upgrade the affected packages to a non-vulnerable version.
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/lexus2016/claude-code-studio)?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