
SutramX
io.github.SutramXv0.1.1Updated Oct 4, 2026
SutramX uptime monitoring: monitors, check results, incidents, status pages and uptime/SLO reports.
Overview
Lets an assistant read and manage SutramX uptime monitoring: monitors, check results, incidents, status pages and uptime/SLO reports.
- What it does
- Connects an assistant to a SutramX workspace through the public API. Read tools cover workspace info, monitor summaries and lists, check history, incidents, status pages, regions and uptime/SLO reports with error budgets. Write tools create, update, pause, resume and run monitors, acknowledge or resolve incidents, add notes, and create or update status pages and their monitor lists. Two delete tools stay hidden unless explicitly enabled.
- When to use it
- Useful when an assistant should triage downtime, answer questions about monitor status and incidents, or produce uptime and SLO reporting from a SutramX workspace. Also fits agents that need to create or adjust monitors and status pages. Read-only mode suits assistants that only need to look.
- Requirements
- Local Node.js 20 or newer process, run via npx or a built entry point. A SutramX workspace API key in SUTRAMX_API_KEY (a Read-only key is recommended for agents). Optional settings: SUTRAMX_READ_ONLY, SUTRAMX_ALLOW_DESTRUCTIVE, SUTRAMX_API_URL, and HTTP mode variables HOST, PORT, MCP_ALLOWED_HOSTS, MCP_ALLOWED_ORIGINS. Network access to the SutramX API.
Installation
In SourceWeft
- Open SutramX in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
SutramX MCP server
A Model Context Protocol server that lets AI agents and assistants (Claude Code, Claude Desktop and other MCP clients) work with your SutramX workspace: list and change monitors, read check results, handle incidents and manage status pages.
It talks to the public SutramX API with a workspace API key, so it can do exactly what that key can do and nothing more.
Tools
Destructive tools are annotated with destructiveHint, so clients that support it ask before running them.
Access modes
Agents can be steered by text they read (prompt injection), so the user, not the agent, chooses which tools the server offers. Tools that are not allowed are not registered at all: the agent never sees them.
- Both settings are off by default, so
sutramx_delete_monitorandsutramx_delete_status_pageare hidden unless deletes are enabled.true,1,yesandon(any case) turn a setting on; anything else leaves it off. - Read-only wins: with read-only on, the delete tools stay hidden even when deletes are enabled.
- HTTP mode: a header can only add to the server's environment, not take away from it.
SUTRAMX_READ_ONLY=trueon the server makes every request read-only whatever the headers say, andSUTRAMX_ALLOW_DESTRUCTIVE=trueon the server enables deletes for every request that is not read-only. Without them, each client chooses with its own headers. - Use read-only mode for assistants that only need to look (triage, reporting), ideally together with a Read-only API key (below).
1. Create an API key
In SutramX, open Settings → API keys → Create API key. Keys start with sk_ and are shown once.
Recommended for agents: a Read-only key. It can call every "no" tool above, and the SutramX API itself refuses every change it attempts (403 READ_ONLY_ACCESS), so a leaked key or a prompt-injected agent cannot create, pause, acknowledge, resolve or delete anything. This holds even if the server's own read-only mode is off; combine both (SUTRAMX_READ_ONLY=true hides the write tools from the agent as well). Use a Standard key only for an agent that really needs to change things; it is enough for every tool above.
2. Build
Node.js 20 or newer is required.
3. Connect a client
Claude Code
Or add it to .mcp.json in a project:
Claude Desktop
Edit claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json, Windows: %APPDATA%\Claude\claude_desktop_config.json):
Restart Claude Desktop. Try: "Which SutramX monitors are down right now, and since when?"
Streamable HTTP (remote or shared)
The endpoint is POST http://127.0.0.1:3333/mcp (stateless, JSON responses). Every request must send its own SutramX key:
so one server can serve several users and workspaces. Connect Claude Code to it with:
When the server listens on loopback (the default, HOST=127.0.0.1), SUTRAMX_API_KEY is used for requests that send no Authorization header (a header that is not a SutramX key is rejected with 401). Browser requests are only accepted from a loopback Origin or one listed in MCP_ALLOWED_ORIGINS. With any other HOST the environment key is ignored, and you should put the server behind HTTPS. Set MCP_ALLOWED_HOSTS (comma-separated host names) to keep DNS-rebinding protection when binding to 0.0.0.0. GET /health returns the server version.
Configuration
Notes
- Plan limits apply exactly as in the dashboard. When a tool returns
ENTITLEMENT_LIMIT_REACHEDorFEATURE_NOT_AVAILABLE, the plan does not allow it. - Billing, team members and API keys cannot be managed with an API key, and so not with this server. Per-monitor alert recipients (
config.notification_emails) can only be set with an API key that has Automation access; other keys get a clear error. - Maintenance windows can be listed but not created, changed or deleted: they silence alerting, so the API makes them owner-only and refuses every API key (
403 WORKSPACE_OWNER_REQUIRED). SLO targets are set in the dashboard;sutramx_uptime_reportreads them. - With a read-only key,
sutramx_whoamireportsread_only: true, and any write tool returnsREAD_ONLY_ACCESSwith a hint telling the agent not to retry. configonsutramx_update_monitorreplaces the whole object. Agents are told to read the monitor first and send the merged config.
Security
SUTRAMX_API_URLmust behttps://(plainhttp://only for loopback); the server refuses to start otherwise, warns whenNODE_TLS_REJECT_UNAUTHORIZED=0, and never follows redirects with a key.- Credentials stored in monitor config (headers such as
Authorization/Cookie, keys named like*token*,*secret*,*password*,*api_key*, and passwords in URLs) are returned as[REDACTED]. Sending[REDACTED]back insutramx_update_monitorkeeps the stored value; it cannot stand for a value that is not stored. - Text that comes from monitored sites or other people (check errors, names, notes) is shown single-line inside
«», without control characters, and the server instructions tell the agent to treat it as data. - Ids are UUIDs, keys and slugs match strict patterns, times must be ISO-8601, so tool arguments cannot change the API path or add query parameters.
- HTTP mode: every request needs its own
Authorization: Bearer sk_...; the env key is only a fallback on loopback. Put a rate limiter in front of a public deployment (the API itself rate-limits per key).
Development
License
MIT, see LICENSE.
Source: README.md at commit 9b2526a
Tools
0Version history
1- v0.1.1LatestOct 4, 2026

