repl-mcp
Enables creation and management of persistent REPL sessions for various languages and shells, with web-based monitoring and session recovery capabilities.
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
takafu
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_sessionCreate a new REPL session with preset or custom configuration. Use displayName to set a custom name that appears in the browser tab title. Returns a webUrl - you MUST display this URL to the user or open it in a browser. Note: If using zsh in custom commands, you may need to manually run 'histchars=…
send_input_to_sessionSend input to a REPL session. Can either wait for prompt return or send input immediately for interactive programs. Output is automatically truncated at ~50,000 characters (~50KB for ASCII, more for multibyte) for responsiveness - use get_full_output for complete large outputs. ALWAYS show the comma…
send_signal_to_sessionSend a signal (like Ctrl+C, Ctrl+Z) to a REPL session process. Can be used even when session is executing to interrupt long-running commands.
list_repl_sessionsList all active REPL sessions. Each session includes a webUrl for browser access. Please show the webUrls to the user so they can open sessions in their browser.
get_session_detailsGet detailed information about a specific session. Includes webUrl for browser access. Please show the webUrl to the user so they can open the session in their browser.
destroy_repl_sessionDestroy an existing REPL session
list_repl_configurationsList all available predefined REPL configurations
set_session_readyMark a session as ready by specifying the detected prompt pattern. Supports regex patterns (e.g., '[\d]+] pry\(main\)> $') or literal strings (e.g., '$ '). Used during LLM-assisted session recovery.
wait_for_sessionWait additional seconds for a session to become ready. Used during LLM-assisted session recovery when more time is needed.
mark_session_failedMark a session as failed with a specific reason. Used during LLM-assisted session recovery when recovery is not possible.
get_full_outputGet the last command's output buffer in chunks. Returns output with pagination info (totalLength, hasMore, nextOffset). Use multiple requests with offset/limit (measured in characters). Default limit is 40,000 characters (~40KB for ASCII, more for multibyte) per request.
get_clean_textGet clean terminal text without ANSI escape codes. Returns either the current line (where cursor is) or full terminal content depending on the fullText parameter.
Disclosed vulnerabilities in this server's declared npm dependencies (via OSV). Whether each is reachable depends on the installed versions.
@modelcontextprotocol/sdk has cross-client data leak via shared server/transport instance reuse
Anthropic's MCP TypeScript SDK has a ReDoS vulnerability
Model Context Protocol (MCP) TypeScript SDK does not enable DNS rebinding protection by default
ws: Uninitialized memory disclosure
REPL_MCP_DEBUG1: Enable verbose debug logging for performance troubleshooting and developmentTool 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 — create_session (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); send_input_to_session (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); send_signal_to_session (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.).
Tests exist
No test files found
Add tests that exercise each declared tool.
Shell command execution
1 child_process call — runs shell commands
Prefer library functions over shell-outs. If you must shell out, ensure all inputs are properly escaped.
Production dependencies are patched
0 critical, 3 high severity in production deps — @modelcontextprotocol/sdk@1.15.0 (high), @modelcontextprotocol/sdk@1.15.0 (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/takafu/repl-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