UpHeron MCP server
com.upheron/upheronupheron.comstreamable-httpv1.2.0Scanned
Uptime and vulnerability monitoring: check sites, manage monitors, incidents and status pages.
59 tools: 6 can delete or cancel things, 2 do something that can't be undone, 30 change things and 21 only read.
- Reads private data · 5
- Reads text others write · 20
- Can send data out · 5
- 3 findings
Tools listed without sign-in. Remote posture
Anlyon alternative
38 of these tools change something. Run those as governed actions.
Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.
Anlyon Radar does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.
Import an OpenAPI spec →Its sign-in does not let Anlyon set up an OAuth client on its own, so there is no link to it here.
What the UpHeron MCP server can do
The UpHeron MCP server is published by upheron.com, a domain the MCP Registry verified. It has 59 tools. It can delete or cancel things (6 tools: delete_on_call_schedule, remove_on_call_override, delete_escalation_policy, delete_maintenance_window and 2 more), do things that cannot be undone, such as sending, paying or publishing (post_incident_update and test_notification_contact) and create or change data (30 tools: check_website, acknowledge_incident, comment_on_incident, open_incident and 26 more). 21 tools only read.
5 tools read private data (resolve_incident, preview_monitor_import, list_notification_contacts and 2 more); 20 tools bring in text other people can write (check_website, get_incident, post_incident_update and 17 more); 5 tools can send data out of your system (check_website, post_incident_update, create_monitor and 2 more). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.
The fixed checks matched 3 findings (annotation-conflict and credentials-mention). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed 2 times since Anlyon Radar first saw them.
Last scanned Nov 12, 2026, version 1.2.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
59 tools
Deletes or cancels 6Claude Code: deny
delete_escalation_policyDeletes a policy; its monitors go back to alerting their own contacts.No findings
Why it is in this group
- The name "delete_escalation_policy" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Deletes a policy; its monitors go back to alerting their own contacts. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
delete_maintenance_windowDeletes a maintenance window, every occurrence of a repeating one included.Reads text others writeNo findings
Why it is in this group
- The name "delete_maintenance_window" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Deletes a maintenance window, every occurrence of a repeating one included. Subscribers told about an occurrence that has not started hear it is cancelled. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
delete_monitorDeletes a monitor and its history.No findings
Why it is in this group
- The name "delete_monitor" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Deletes a monitor and its history. Frees its plan slot.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
delete_on_call_scheduleDeletes a schedule no escalation policy uses.No findings
Why it is in this group
- The name "delete_on_call_schedule" says "delete": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Deletes a schedule no escalation policy uses. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
remove_on_call_overrideRemoves an override from a schedule (its id is in the schedule's overrides).No findings
Why it is in this group
- The name "remove_on_call_override" says "remove": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Removes an override from a schedule (its id is in the schedule's overrides). Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
remove_softwareRemoves an entry from get_vulnerabilities' software list.No findings
Why it is in this group
- The name "remove_software" says "remove": a word for removing or ending something.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Removes an entry from get_vulnerabilities' software list. Software the scan found is ignored instead, so the next scan does not add it back. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
Can't be undone 2Claude Code: ask
post_incident_updatePublishes an update on every published status page that shows the incident's monitor and in their RSS and Atom feeds, and on Pro plans emails those pages' subscribers.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "post_incident_update" says "post": a word for a change.
- The description's first verb is "publish": a word for a change that another call does not undo.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe name "post_incident_update" says "post": it sends or publishes something to other people.
Description, as the server wrote it Publishes an update on every published status page that shows the incident's monitor and in their RSS and Atom feeds, and on Pro plans emails those pages' subscribers. States: investigating, identified, monitoring, resolved. Write for the page's visitors. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
test_notification_contactSends a contact a test alert, at most 5 a minute for a workspace.Reads text others writeCan send data out1 finding
Why it is in this group
- The description's first verb is "send": a word for sending, and what is sent cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe description's first verb is "send": it sends or publishes something to other people.
Findings
- Info
The description mentions "credentials".
What this check looks forMatched text
description: "...ays in words what the provider refused (credentials, a trial account's unverified number, t..."
Description, as the server wrote it Sends a contact a test alert, at most 5 a minute for a workspace. For a phone contact it is a real text or call through the user's account, which counts toward its 6 texts or 3 calls an hour; an error says in words what the provider refused (credentials, a trial account's unverified number, the sender).
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
Changes things 30Claude Code: ask
accept_page_changeMakes the page as it is now the new baseline of a monitor with content_watch, after get_website_changes shows a change you or the site's owner made on purpose.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Makes the page as it is now the new baseline of a monitor with content_watch, after get_website_changes shows a change you or the site's owner made on purpose. Later looks compare with it. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
acknowledge_incidentMarks an open incident as being handled: its repeat alerts and escalation stop.No findings
Why it is in this group
- The description's first verb is "mark": a word for a change.
Description, as the server wrote it Marks an open incident as being handled: its repeat alerts and escalation stop. Everyone alerted still hears when it recovers. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
add_notification_contactAdds an email address, chat channel, paging service, Jira project, webhook or phone number to alert on outages and recoveries.Reads text others write1 finding
Why it is in this group
- The name "add_notification_contact" says "add": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Info
The description mentions "credentials".
What this check looks forMatched text
description: "...(API key) text. The user gives you the credentials and the sender, one of the account's nu..."
Description, as the server wrote it Adds an email address, chat channel, paging service, Jira project, webhook or phone number to alert on outages and recoveries. Email contacts must open a confirmation link before they receive alerts. A phone number texts or calls through the user's own account with a provider, on every plan, which bills them: twilio (Account SID, Auth Token) and plivo (Auth ID, Auth Token) text or call, vonage (API key and secret) and telnyx (API key) text. The user gives you the credentials and the sender, one of the account's numbers or for texts a sender ID of up to 11 letters and digits (not in the US or Canada). It needs no code: check it with test_notification_contact. sms and voice, through UpHeron's own Twilio account, are no longer added.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
add_on_call_overrideBusiness plans: puts a contact on call in a schedule from starts_at to ends_at (up to 90 days), instead of the rotation.No findings
Why it is in this group
- The name "add_on_call_override" says "add": a word for a change.
Description, as the server wrote it Business plans: puts a contact on call in a schedule from starts_at to ends_at (up to 90 days), instead of the rotation. Overrides may not overlap. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
add_softwareAdds one piece of software the website runs, by name (Debian, nginx, WordPress, Next.js…), package (npm:lodash, composer:vendor/package, pypi:django, wp-plugin:slug) or NVD product…No findings
Why it is in this group
- The name "add_software" says "add": a word for a change.
Description, as the server wrote it Adds one piece of software the website runs, by name (Debian, nginx, WordPress, Next.js…), package (npm:lodash, composer:vendor/package, pypi:django, wp-plugin:slug) or NVD product (cpe:vendor:product), with its version if known. Replaces a detected or imported entry for the same product. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
change_maintenance_occurrenceSkips (skipped: true), moves (starts_at and ends_at) or changes the message of one occurrence of a repeating maintenance window, leaving the others as they are; restore: true undoe…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Skips (skipped: true), moves (starts_at and ends_at) or changes the message of one occurrence of a repeating maintenance window, leaving the others as they are; restore: true undoes it. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
check_websiteChecks whether a public website responds right now, from the monitoring service, without an account.Reads text others writeCan send data outNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Checks whether a public website responds right now, from the monitoring service, without an account. Returns status, HTTP code and timings. Free; limited to a few checks a minute.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: true
comment_on_incidentAdds a comment to the incident's timeline, seen only by the workspace (Pro plans).No findings
Why it is in this group
- The name "comment_on_incident" says "comment": a word for a change.
Description, as the server wrote it Adds a comment to the incident's timeline, seen only by the workspace (Pro plans). Use it to record what you found or did. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
confirm_notification_contactConfirms an sms or voice contact, added before they were retired, with the 6-digit code it was texted or told in a call.No findings
Why it is in this group
- The input takes a statement or command to run (input.code), which can change anything it can reach.
Description, as the server wrote it Confirms an sms or voice contact, added before they were retired, with the 6-digit code it was texted or told in a call. The code works for 24 hours and 5 tries. Ask the phone's owner for it; never guess.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
create_escalation_policyBusiness plans: ordered steps, each alerting contacts and whoever is on call in schedules, then waiting for an acknowledgement before the next.No findings
Why it is in this group
- The name "create_escalation_policy" says "create": a word for a change.
Description, as the server wrote it Business plans: ordered steps, each alerting contacts and whoever is on call in schedules, then waiting for an acknowledgement before the next. Monitors use it with update_monitor's escalation_policy_id. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
create_maintenance_windowSchedules maintenance for one monitor or all: no incidents or alerts while it lasts (suppress_alerts), and with announce the status pages that show it list it and tell their subscr…Reads text others writeNo findings
Why it is in this group
- The name "create_maintenance_window" says "create": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Schedules maintenance for one monitor or all: no incidents or alerts while it lasts (suppress_alerts), and with announce the status pages that show it list it and tell their subscribers. With recurrence it repeats daily, weekly on chosen days or monthly, in the workspace's time zone (get_account), starting with starts_at to ends_at. Every plan. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
create_monitorCreates a monitor in a free slot of the workspace's plan.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_monitor" says "create": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Creates a monitor in a free slot of the workspace's plan. Premium monitors (multi-step transactions and code repositories) have slots of their own, on Pro and Business only. When every slot is in use the call fails with monitor_limit_reached: the user can upgrade at https://upheron.com/pricing or delete a monitor. Needs a write token.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
create_on_call_scheduleBusiness plans: a rotation through notification contacts (list_notification_contacts), handing off daily or weekly at a time in a time zone.No findings
Why it is in this group
- The name "create_on_call_schedule" says "create": a word for a change.
Description, as the server wrote it Business plans: a rotation through notification contacts (list_notification_contacts), handing off daily or weekly at a time in a time zone. The first member is on call from now. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
import_monitorsCreates the monitors the user chose from preview_monitor_import (send their `monitor` objects).No findings
Why it is in this group
- The name "import_monitors" says "import": a word for a change.
Description, as the server wrote it Creates the monitors the user chose from preview_monitor_import (send their `monitor` objects). Each takes a plan slot like create_monitor; ones already monitored are skipped. Imported heartbeats get new URLs, returned once as heartbeat_url: tell the user to point their scheduled jobs at them. Needs a write token. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
import_softwareTells the monitor what software the website and its server run, so their vulnerabilities are checked.No findings
Why it is in this group
- The name "import_software" says "import": a word for a change.
Description, as the server wrote it Tells the monitor what software the website and its server run, so their vulnerabilities are checked. Send a file's contents (package.json, package-lock.json, yarn.lock, pnpm-lock.yaml, bun.lock, mix.lock, composer.json/.lock, requirements.txt, Pipfile.lock, poetry.lock, Gemfile.lock, go.mod, Cargo.lock, pom.xml, .csproj, Dockerfile, docker-compose.yml, .nvmrc, .bun-version, .tool-versions, mise.toml, railpack.json) or a plain list with one item per line as "name version", for example "Debian 12", "nginx 1.24.0", "PHP 8.2.12", "wp-plugin:woocommerce 9.3.3", "npm:lodash 4.17.21". If you can run commands on the server, `cat /etc/os-release`, `nginx -v`, `php -v` and the like give exact versions. Importing the same filename again replaces what it added. Lines that name nothing known are returned as unrecognized. Format and supported files: https://upheron.com/docs/software. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
mark_vulnerability_fixedRecords that the software behind a finding from get_vulnerabilities was upgraded: to `version`, or by default to the first version that fixes it (the finding's fix_version).No findings
Why it is in this group
- The name "mark_vulnerability_fixed" says "mark": a word for a change.
Description, as the server wrote it Records that the software behind a finding from get_vulnerabilities was upgraded: to `version`, or by default to the first version that fixes it (the finding's fix_version). The monitor is checked again; findings that no longer apply disappear, and one that still applies stays, marked with the version. Only records the version; it does not change the server. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
open_incidentOpens an incident for what the checks cannot see: a slow service, a problem a customer reported.Reads text others writeNo findings
Why it is in this group
- The name "open_incident" says "open": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Opens an incident for what the checks cannot see: a slow service, a problem a customer reported. With a message it publishes the first update on the status pages that show its monitors (and emails their subscribers on Pro plans); alert pages the monitors' contacts once. It stays open until resolve_incident and never counts as downtime. Every plan, at most 20 open. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
pause_monitorStops checking a monitor.No findings
Why it is in this group
- The description's first verb is "stop": a word for a change.
Description, as the server wrote it Stops checking a monitor. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
resolve_incidentResolves an incident opened with open_incident; the contacts it alerted hear so.Reads private dataReads text others write1 finding
Why it is in this group
- The name "resolve_incident" says "resolve": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
What it can reach
- private-dataThe description's first sentence says "resolve" and "contact": it reads contacts.
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The server marks "resolve_incident" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "resolve_incident" says "resolve": a word for reading.
Description, as the server wrote it Resolves an incident opened with open_incident; the contacts it alerted hear so. A message is published as the last update on its status pages. Incidents opened by checks resolve with their checks. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
resume_monitorStarts checking a paused monitor again.No findings
Why it is in this group
- The description's first verb is "start": a word for a change.
Description, as the server wrote it Starts checking a paused monitor again. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
scan_portsScans the verified host's common TCP ports now instead of waiting for the weekly scan, at most once an hour.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Scans the verified host's common TCP ports now instead of waiting for the weekly scan, at most once an hour. Takes a minute or two; then read get_ports. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
scan_website_softwareScans the monitor's website for the software it runs and checks it against the vulnerability sources now, instead of waiting for the next scan.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Scans the monitor's website for the software it runs and checks it against the vulnerability sources now, instead of waiting for the next scan. Takes up to a minute; then read get_vulnerabilities. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
update_escalation_policyBusiness plans.No findings
Why it is in this group
- The name "update_escalation_policy" says "update": a word for a change.
Description, as the server wrote it Business plans. Fields left out keep their value; steps replaces them all. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
update_incidentChanges the title status pages show (public_title) or the root cause (Pro plans) of any incident, and the title, severity and monitors of one opened by hand.No findings
Why it is in this group
- The name "update_incident" says "update": a word for a change.
Description, as the server wrote it Changes the title status pages show (public_title) or the root cause (Pro plans) of any incident, and the title, severity and monitors of one opened by hand. Fields left out keep their value. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
update_maintenance_windowChanges a maintenance window or every occurrence of a repeating one.Reads text others writeNo findings
Why it is in this group
- The name "update_maintenance_window" says "update": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Changes a maintenance window or every occurrence of a repeating one. Fields left out keep their value; recurrence null makes it a single window. Changing the times or recurrence undoes skipped and moved occurrences. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
update_monitorChanges a monitor's settings, within the plan's interval and location limits.Can send data outNo findings
Why it is in this group
- The name "update_monitor" says "update": a word for a change.
What it can reach
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Changes a monitor's settings, within the plan's interval and location limits. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
update_on_call_scheduleBusiness plans.No findings
Why it is in this group
- The name "update_on_call_schedule" says "update": a word for a change.
Description, as the server wrote it Business plans. Fields left out keep their value; changing the members, rotation, handoff or time zone starts the rotation over. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
update_status_page_componentsReplaces the monitors a status page shows, in order, each optionally in a named group ("API", "EU region"); a group sits where its first monitor is and shows their worst state.No findings
Why it is in this group
- The name "update_status_page_components" says "update": a word for a change.
Description, as the server wrote it Replaces the monitors a status page shows, in order, each optionally in a named group ("API", "EU region"); a group sits where its first monitor is and shows their worst state. Every plan. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
verify_hostLooks for the proof of each of the host's addresses from get_ports: the file served at the address, or its reverse DNS (PTR) under the host resolving back to it with the TXT record…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Looks for the proof of each of the host's addresses from get_ports: the file served at the address, or its reverse DNS (PTR) under the host resolving back to it with the TXT record in place. The TXT record alone proves no address. The proof must stay in place, since every scan checks it again; only covered addresses are scanned. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
write_postmortemWrites a resolved incident's post-mortem: summary, impact, root cause and follow-ups (one per line), on Pro plans.Reads text others writeNo findings
Why it is in this group
- The name "write_postmortem" says "write": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Writes a resolved incident's post-mortem: summary, impact, root cause and follow-ups (one per line), on Pro plans. It stays in the workspace until published: true shows it under the incident on the status pages that show it and in their feeds; false takes it off. Publish only when the user asks. Free.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
Only reads 21Show 21 read-only tools
get_accountPlan and its monitor slots (standard in plan_monitors, premium in premium_monitors with their kinds), and the workspace's time zone (workspace.time_zone), which report days follow.No findings
Why it is in this group
- The name "get_account" says "get": a word for reading.
Description, as the server wrote it Plan and its monitor slots (standard in plan_monitors, premium in premium_monitors with their kinds), and the workspace's time zone (workspace.time_zone), which report days follow. Call this first to know whether the plan has room for another monitor.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_audit_logWho changed what in the workspace, newest first: people, API tokens and agents, with the changed fields and IP (Business plan).No findings
Why it is in this group
- The name "get_audit_log" says "get": a word for reading.
Description, as the server wrote it Who changed what in the workspace, newest first: people, API tokens and agents, with the changed fields and IP (Business plan). Filter by resource_type (monitor, alert_channel, status_page, member, api_token…), action (monitor.updated, or a prefix like monitor.), from and to (YYYY-MM-DD); pass next as before for older entries. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_domain_watchFor an http or api monitor on Pro and Business plans: the website's registered domain, its subdomains from certificate transparency logs (crt.sh, which can be days or weeks behind)…Reads private dataReads text others writeNo findings
Why it is in this group
- The name "get_domain_watch" says "get": a word for reading.
What it can reach
- private-dataThe description's first sentence says "mail": it reads mail.
- untrusted-inputThe description's first sentence says "http": it reads web pages, which anyone can write.
Description, as the server wrote it For an http or api monitor on Pro and Business plans: the website's registered domain, its subdomains from certificate transparency logs (crt.sh, which can be days or weeks behind) with where each points and any CNAME to a service that reports no site there (subdomain_alerts), and the registered typo and look-alike spellings of the domain with their addresses, mail servers and registration dates (lookalike_alerts). Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_error_budgetsEvery monitor with an availability objective (slo_target over slo_window_days, set with update_monitor): its availability over the rolling window, the error budget in seconds, how …No findings
Why it is in this group
- The name "get_error_budgets" says "get": a word for reading.
Description, as the server wrote it Every monitor with an availability objective (slo_target over slo_window_days, set with update_monitor): its availability over the rolling window, the error budget in seconds, how much is used and left, the last hour's burn rate and whether it burns fast, least budget left first. Every plan. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_incidentOne incident with its timeline (detected, acknowledged, comments, status page updates, recovered), root cause and post-mortem, and the status pages its updates go to with their sub…Reads text others writeNo findings
Why it is in this group
- The name "get_incident" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "comment": it reads comments, which other people can write.
Description, as the server wrote it One incident with its timeline (detected, acknowledged, comments, status page updates, recovered), root cause and post-mortem, and the status pages its updates go to with their subscriber counts. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_monitorOne monitor with uptime for 24 hours, 7 days and 30 days, its last 90 days, incident totals, its latest results with errors, recent incidents, and route traces from failing locatio…No findings
Why it is in this group
- The name "get_monitor" says "get": a word for reading.
Description, as the server wrote it One monitor with uptime for 24 hours, 7 days and 30 days, its last 90 days, incident totals, its latest results with errors, recent incidents, and route traces from failing locations (Pro and Business: each hop's loss, round trips and jitter over every round). Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_monitor_resultsA monitor's most recent check results, newest first.No findings
Why it is in this group
- The name "get_monitor_results" says "get": a word for reading.
Description, as the server wrote it A monitor's most recent check results, newest first. Older results may be merged: repeats checks with the same outcome up to last_checked_at. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_portsFor a monitor with port_scan on (Business plans): which of its host's addresses are verified and how to verify the rest (a file each address serves, or reverse DNS under the host w…Reads private dataNo findings
Why it is in this group
- The name "get_ports" says "get": a word for reading.
What it can reach
- private-dataThe description's first sentence says "file": it reads files.
Description, as the server wrote it For a monitor with port_scan on (Business plans): which of its host's addresses are verified and how to verify the rest (a file each address serves, or reverse DNS under the host with a TXT record), the ports the last weekly scan found open, flagged where the service usually should not be public, and the recent scans. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_sla_reportBusiness plans: availability against an SLA target for a range of days in the workspace's time zone (time_zone in get_account; by default this month), with the downtime budget used…No findings
Why it is in this group
- The name "get_sla_report" says "get": a word for reading.
Description, as the server wrote it Business plans: availability against an SLA target for a range of days in the workspace's time zone (time_zone in get_account; by default this month), with the downtime budget used and left, incidents, MTTR, MTBF and the longest outage, per monitor (lowest availability first) and overall. The dashboard and the API also give it as a PDF. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_uptime_historyA monitor's daily uptime, kept for as long as it exists: checks, successful, failed and average response time per UTC day, oldest first.No findings
Why it is in this group
- The name "get_uptime_history" says "get": a word for reading.
Description, as the server wrote it A monitor's daily uptime, kept for as long as it exists: checks, successful, failed and average response time per UTC day, oldest first. By default from the first check to today. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_vulnerabilitiesFor an http or api monitor with vulnerability_alerts on: the vulnerabilities and unsupported releases found in the website's software, actively exploited and most severe first, wit…Reads text others writeNo findings
Why it is in this group
- The name "get_vulnerabilities" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "http": it reads web pages, which anyone can write.
Description, as the server wrote it For an http or api monitor with vulnerability_alerts on: the vulnerabilities and unsupported releases found in the website's software, actively exploited and most severe first, with the version that fixes each, and the software they were matched on (found by scanning the website, imported, or added). Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_website_changesFor a website or API monitor: its security headers grade with what each missing or weak header does and how to fix it, its DNS records (A, AAAA, CNAME, NS, MX, SPF, DMARC) and the …Reads text others writeNo findings
Why it is in this group
- The name "get_website_changes" says "get": a word for reading.
What it can reach
- untrusted-inputThe name "get_website_changes" says "get" and "website": it reads web pages, which anyone can write.
- untrusted-inputThe description's first sentence says "website": it reads web pages, which anyone can write.
Description, as the server wrote it For a website or API monitor: its security headers grade with what each missing or weak header does and how to fix it, its DNS records (A, AAAA, CNAME, NS, MX, SPF, DMARC) and the networks its addresses belong to, its page fingerprint (Business: title, external script hosts, and the latest change waiting to be accepted) and the recent header and DNS changes. Settings are header_watch, dns_watch, dns_address_changes, content_watch and content_threshold on update_monitor. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_escalation_policiesBusiness plans: escalation policies, whose id a monitor's escalation_policy_id takes, and their steps.No findings
Why it is in this group
- The name "list_escalation_policies" says "list": a word for reading.
Description, as the server wrote it Business plans: escalation policies, whose id a monitor's escalation_policy_id takes, and their steps. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_incidentsOutages, newest first, with their status (open, acknowledged, resolved).No findings
Why it is in this group
- The name "list_incidents" says "list": a word for reading.
Description, as the server wrote it Outages, newest first, with their status (open, acknowledged, resolved). Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_locationsLocation codes you can pass in a monitor's `locations`.No findings
Why it is in this group
- The name "list_locations" says "list": a word for reading.
Description, as the server wrote it Location codes you can pass in a monitor's `locations`.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_maintenance_windowsMaintenance windows, latest first: times when failures open no incidents and send no alerts, for one monitor or all.No findings
Why it is in this group
- The name "list_maintenance_windows" says "list": a word for reading.
Description, as the server wrote it Maintenance windows, latest first: times when failures open no incidents and send no alerts, for one monitor or all. A repeating window has its recurrence in words (repeats), its next occurrences and the ones skipped or moved. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_monitorsEvery monitor with its current status (pending, up, down, degraded), and interval.No findings
Why it is in this group
- The name "list_monitors" says "list": a word for reading.
Description, as the server wrote it Every monitor with its current status (pending, up, down, degraded), and interval. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_notification_contactsContacts that receive alerts, whether each is confirmed, and for phone contacts their latest text or call and whether it was delivered.Reads private dataNo findings
Why it is in this group
- The name "list_notification_contacts" says "list": a word for reading.
What it can reach
- private-dataThe name "list_notification_contacts" says "list" and "contact": it reads contacts.
- private-dataThe description's first sentence says "contact": it reads contacts.
Description, as the server wrote it Contacts that receive alerts, whether each is confirmed, and for phone contacts their latest text or call and whether it was delivered.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_on_call_schedulesBusiness plans: on-call schedules with who is on call now, until when, and who is next.No findings
Why it is in this group
- The name "list_on_call_schedules" says "list": a word for reading.
Description, as the server wrote it Business plans: on-call schedules with who is on call now, until when, and who is next. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
list_status_pagesThe workspace's status pages: their address (/status/{slug}), the monitors they show with the names visitors see, and their named groups of monitors in order.No findings
Why it is in this group
- The name "list_status_pages" says "list": a word for reading.
Description, as the server wrote it The workspace's status pages: their address (/status/{slug}), the monitors they show with the names visitors see, and their named groups of monitors in order. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
preview_monitor_importReads the monitors of another service: UptimeRobot (a read-only API key, or its bulk upload CSV), Better Stack (a team-based API token), Uptime Kuma (its JSON backup) or a CSV file…Reads private dataReads text others writeNo findings
Why it is in this group
- The name "preview_monitor_import" says "preview": a word for reading.
What it can reach
- private-dataThe description's first sentence says "read" and "file": it reads files.
- private-dataThe input names a file (input.file), and the tool reads: it reads files.
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe description's first sentence says "read" and "http": it reads web pages, which anyone can write.
Description, as the server wrote it Reads the monitors of another service: UptimeRobot (a read-only API key, or its bulk upload CSV), Better Stack (a team-based API token), Uptime Kuma (its JSON backup) or a CSV file in UpHeron's format (https://upheron.com/docs/import). Creates nothing: says which would be created and which skipped and why (no UpHeron check for it, already monitored, beyond the plan's monitors, needs Pro), with the UpHeron monitor each maps to. Show the user the list before import_monitors. The key is used once and never stored. Needs a write token. Free.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
500 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 5 outbound tools breaks every one.
- Reads private data5
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write20
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out5
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__upheron__check_website",
"mcp__upheron__create_monitor",
"mcp__upheron__post_incident_update",
"mcp__upheron__test_notification_contact",
"mcp__upheron__update_monitor"
]
}
}Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 19, 2026, 17:21 UTC, unchanged since Nov 19, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 38
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://upheron.com/.well-known/oauth-protected-resource
- Authorization servers
- https://upheron.com
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- no
- Client ID metadata documents
- no
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 38 tools that change things (6 destructive, 2 irreversible, 30 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://upheron.com/mcp: tools/list returned 59 tools to an anonymous client, including delete_on_call_schedule (destructive), remove_on_call_override (destructive), delete_escalation_policy (destructive), delete_maintenance_window (destructive), delete_monitor (destructive) and 33 more
Posture changes
- Nov 19, 2026
- Tools that change things, listed without credentials: 31 to 38
Findings
Matches of fixed checks, with the text they matched.
- Warnresolve_incident
The server marks "resolve_incident" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "resolve_incident" says "resolve": a word for reading.
The description mentions "credentials".
What this check looks forMatched text
description: "...(API key) text. The user gives you the credentials and the sender, one of the account's nu..."
The description mentions "credentials".
What this check looks forMatched text
description: "...ays in words what the provider refused (credentials, a trial account's unverified number, t..."
History
Changes to its tool definitions and to its remote endpoint, newest first.
PostureRemote posture changed. Tools that change things, listed without credentials: 31 -> 38.View diff
- Tools that change things, listed without credentials: 31 to 38
Tools13 added, 2 changed.+13 added~2 changedView diff
- Added:accept_page_changeChanges things
- Added:change_maintenance_occurrenceChanges things
- Added:create_maintenance_windowChanges things
- Added:delete_maintenance_windowDeletes or cancels
- Added:get_domain_watchOnly reads
- Added:get_error_budgetsOnly reads
- Added:get_website_changesOnly reads
- Added:import_monitorsChanges things
- Added:list_maintenance_windowsOnly reads
- Added:list_status_pagesOnly reads
- Added:preview_monitor_importOnly reads
- Added:update_maintenance_windowChanges things
- Added:update_status_page_componentsChanges things
- Changed:create_monitorChanges thingsinputSchema
- Changed:update_monitorChanges thingsinputSchema
Tools18 changed.~18 changedView diff
- Changed:add_notification_contactChanges thingsinputSchema
- Changed:add_on_call_overrideChanges thingsinputSchema
- Changed:comment_on_incidentChanges thingsinputSchema
- Changed:confirm_notification_contactChanges thingsinputSchema
- Changed:create_escalation_policyChanges thingsinputSchema
- Changed:create_monitorChanges thingsinputSchema
- Changed:create_on_call_scheduleChanges thingsinputSchema
- Changed:get_audit_logOnly readsinputSchema
- Changed:get_monitor_resultsOnly readsinputSchema
- Changed:open_incidentChanges thingsinputSchema
- Changed:post_incident_updateCan't be undoneinputSchema
- Changed:remove_on_call_overrideDeletes or cancelsinputSchema
- Changed:resolve_incidentChanges thingsinputSchema
- Changed:update_escalation_policyChanges thingsinputSchema
- Changed:update_incidentChanges thingsinputSchema
- Changed:update_monitorChanges thingsinputSchema
- Changed:update_on_call_scheduleChanges thingsinputSchema
- Changed:write_postmortemChanges thingsinputSchema
Install
Pinned to version 1.2.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http upheron https://upheron.com/mcpclaude mcp add --transport http upheron https://upheron.com/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__upheron__get_account",
"mcp__upheron__list_locations",
"mcp__upheron__list_monitors",
"mcp__upheron__get_monitor",
"mcp__upheron__get_monitor_results",
"mcp__upheron__get_uptime_history",
"mcp__upheron__get_sla_report",
"mcp__upheron__get_error_budgets",
"mcp__upheron__list_incidents",
"mcp__upheron__list_escalation_policies",
"mcp__upheron__list_on_call_schedules",
"mcp__upheron__get_incident",
"mcp__upheron__list_status_pages",
"mcp__upheron__list_maintenance_windows",
"mcp__upheron__get_audit_log",
"mcp__upheron__preview_monitor_import",
"mcp__upheron__list_notification_contacts",
"mcp__upheron__get_vulnerabilities",
"mcp__upheron__get_ports",
"mcp__upheron__get_website_changes",
"mcp__upheron__get_domain_watch"
],
"ask": [
"mcp__upheron__check_website",
"mcp__upheron__acknowledge_incident",
"mcp__upheron__post_incident_update",
"mcp__upheron__comment_on_incident",
"mcp__upheron__open_incident",
"mcp__upheron__update_incident",
"mcp__upheron__resolve_incident",
"mcp__upheron__write_postmortem",
"mcp__upheron__update_status_page_components",
"mcp__upheron__create_on_call_schedule",
"mcp__upheron__update_on_call_schedule",
"mcp__upheron__add_on_call_override",
"mcp__upheron__create_escalation_policy",
"mcp__upheron__update_escalation_policy",
"mcp__upheron__create_maintenance_window",
"mcp__upheron__update_maintenance_window",
"mcp__upheron__change_maintenance_occurrence",
"mcp__upheron__create_monitor",
"mcp__upheron__import_monitors",
"mcp__upheron__update_monitor",
"mcp__upheron__pause_monitor",
"mcp__upheron__resume_monitor",
"mcp__upheron__add_notification_contact",
"mcp__upheron__confirm_notification_contact",
"mcp__upheron__test_notification_contact",
"mcp__upheron__import_software",
"mcp__upheron__add_software",
"mcp__upheron__mark_vulnerability_fixed",
"mcp__upheron__verify_host",
"mcp__upheron__accept_page_change",
"mcp__upheron__scan_ports",
"mcp__upheron__scan_website_software"
],
"deny": [
"mcp__upheron__delete_on_call_schedule",
"mcp__upheron__remove_on_call_override",
"mcp__upheron__delete_escalation_policy",
"mcp__upheron__delete_maintenance_window",
"mcp__upheron__delete_monitor",
"mcp__upheron__remove_software"
]
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as upheron. Review it before you use it.
Capability tags behind 23 rules
- mcp__upheron__check_website untrusted-input, outbound
- mcp__upheron__get_incident untrusted-input
- mcp__upheron__post_incident_update untrusted-input, outbound
- mcp__upheron__open_incident untrusted-input
- mcp__upheron__resolve_incident private-data, untrusted-input
- mcp__upheron__write_postmortem untrusted-input
- mcp__upheron__create_maintenance_window untrusted-input
- mcp__upheron__update_maintenance_window untrusted-input
- mcp__upheron__change_maintenance_occurrence untrusted-input
- mcp__upheron__delete_maintenance_window untrusted-input
- mcp__upheron__create_monitor untrusted-input, outbound
- mcp__upheron__preview_monitor_import private-data, untrusted-input
- mcp__upheron__update_monitor outbound
- mcp__upheron__add_notification_contact untrusted-input
- mcp__upheron__list_notification_contacts private-data
- mcp__upheron__test_notification_contact untrusted-input, outbound
- mcp__upheron__get_vulnerabilities untrusted-input
- mcp__upheron__get_ports private-data
- mcp__upheron__verify_host untrusted-input
- mcp__upheron__get_website_changes untrusted-input
- mcp__upheron__scan_ports untrusted-input
- mcp__upheron__get_domain_watch private-data, untrusted-input
- mcp__upheron__scan_website_software untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/com.upheron/upheron/verify?hash=c4b1d0e4f3c00d1c40a7eeef2aee7640d3a1c84d388008b3295a3b68b1263ca9" | jq -e '.changed == false'Pinned hash c4b1d0e4f3c0, from Nov 12, 2026, 12:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.upheron/upheron/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.upheron/upheron)Source: MCP Registry. Synced Nov 12, 2026, 12:17 UTC.
UpHeron MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the UpHeron MCP server safe?
- No scan can promise that. What its definitions show: 59 tools, 38 of which can change, send or delete something, and 3 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the UpHeron MCP server delete data?
- Yes. 6 tools can delete, cancel or revoke something: delete_on_call_schedule, remove_on_call_override, delete_escalation_policy, delete_maintenance_window, delete_monitor and remove_software.
Can the UpHeron MCP server do things that cannot be undone?
- Yes. 2 tools can send, pay, publish or merge: post_incident_update and test_notification_contact.
What tools does the UpHeron MCP server have?
- 59 tools: 6 destructive, 2 irreversible, 30 write, 21 read. delete_on_call_schedule, remove_on_call_override, delete_escalation_policy, delete_maintenance_window, delete_monitor, remove_software, post_incident_update, test_notification_contact and 51 more.
Does the UpHeron MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the UpHeron MCP server send data to outside services?
- 5 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): check_website, post_incident_update, create_monitor, update_monitor and test_notification_contact.
Has the UpHeron MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 12, 2026. Every change is on this page and in its Atom feed.