Skip to content

devops-status-mcp-server

io.github.cyanheads/devops-status-mcp-servercyanheadsstreamable-httpv0.11.2Scanned

Vendor status pages, TLS cert inspection, DNS propagation checks, and incident-response playbooks.

7 tools: 1 change things and 6 only read.

Tools: 0 delete, 0 irreversible, 1 write, 6 read.
  • Reads private data · 0
  • Reads text others write · 5
  • Can send data out · 0
  • 4 findings

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here changes things. Govern it.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the devops-status-mcp-server can do

The devops-status-mcp-server is published by the GitHub account cyanheads. It has 7 tools. It can create or change data (1 tool: devops_watch_stack), but cannot delete data. 6 tools only read.

5 tools bring in text other people can write (devops_status_check, devops_get_incidents, devops_watch_stack and 2 more).

The fixed checks matched 4 findings (credentials-mention and long-description). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in, and its tools changed 3 times since Anlyon Radar first saw them.

Last scanned Nov 16, 2026, version 0.11.2.

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.

7 tools

Changes things 1Claude Code: ask
  • devops_watch_stackCheck the health of a named vendor stack — a saved list of vendors representing your infrastructure dependencies.Reads text others write1 finding

    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.

    Findings

    • Info

      The input schema at properties.vendors.description mentions "credentials".

      Matched text

      inputSchema.properties.vendors.description: "...ead and saved at their origin and path (credentials, query, and fragment dropped). Most oth..."

      What this check looks for
    Description, as the server wrote it
    Check the health of a named vendor stack — a saved list of vendors representing your infrastructure dependencies. On the first call, provide vendors to define the stack; subsequent calls can omit vendors to reuse the saved list. Stacks are saved per authenticated caller, or per local server process; a shared deployment that does not authenticate callers saves nothing, so there vendors is required on every call and the response says so. Up to 20 vendors are saved per stack and checked per call; use several stack names for a larger set. Returns a unified health snapshot with an aggregate rollup plus per-vendor detail. A vendor that cannot be resolved or reached is reported in its own row and left out of the saved stack, so one bad entry never discards the sweep. Every vendor with an active problem gets a pre-filled devops_suggest_action call in nextToolSuggestions. Ideal for morning status checks or pre-deploy sweeps. Multiple stacks can coexist (e.g., "production", "staging").

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
Only reads 6Show 6 read-only tools
  • devops_check_certsInspect SSL/TLS certificate health for one or more domains by performing a real TLS handshake.Reads text others writeNo findings

    Why it is in this group

    • The description's first verb is "inspect": a word for reading.

    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
    Inspect SSL/TLS certificate health for one or more domains by performing a real TLS handshake. Works for any internet-accessible domain — no vendor registry required. Reports days to expiry (flagged at < 30 days warning and < 7 days critical), certificate subject and SANs, issuer, hostname coverage, chain-trust verification, TLS protocol version negotiated (flags TLS 1.0/1.1 as insecure — a server offering only those is still inspected), cipher suite, and HSTS presence when the port answers HTTP. The handshake completes even for a certificate clients would reject, so a broken certificate is reported rather than hidden behind a connection error: a hostname mismatch surfaces in cert.hostname_verification_error and a chain-trust failure (self-signed, untrusted root) in cert.authorization_error, both status "critical". Non-HTTP TLS ports (IMAPS 993, SMTPS 465, DNS over TLS 853) are graded the same way, with HSTS reported as not checked. If a domain fails to connect at all, check devops_check_dns first — the name may not resolve. When a certificate fails to issue or renew, run devops_check_dns with record_types ["CAA"] to see which certificate authorities the domain allows.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • devops_check_dnsResolve DNS records for one or more domains across multiple public resolvers and compare what each resolver returned.Reads text others writeNo findings

    Why it is in this group

    • The description's first verb is "resolve": a word for reading.

    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
    Resolve DNS records for one or more domains across multiple public resolvers and compare what each resolver returned. Works for any domain — no vendor registry required. Reports records found (A/AAAA/CNAME/MX/TXT/NS/CAA/SOA), resolution latency per resolver, and a typed outcome per resolver and record type so "the domain does not exist" (nxdomain), "the resolver could not answer" (servfail), and "no record of this type" (nodata) stay distinguishable. Resolver disagreements are reported without asserting a cause: partial_resolution (some resolvers answered, others returned nothing) points at a real propagation or resolver problem, while value_variation (every resolver answered with different values) is the normal steady state for anycast and geo-steered domains. Pair with devops_check_certs when a domain resolves but TLS to it is failing.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • devops_get_incidentsFetch incident history and scheduled maintenance windows for a vendor.Reads text others write2 findings

    Why it is in this group

    • The name "devops_get_incidents" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Info

      The input schema at properties.vendor.description mentions "credentials".

      Matched text

      inputSchema.properties.vendor.description: "...base URL, read at its origin and path (credentials, query, and fragment dropped). Most oth..."

      What this check looks for
    • Info

      The description is 2,060 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...result, or a history read that stopped short of since, explains itself in notice."

      What this check looks for
    Description, as the server wrote it
    Fetch incident history and scheduled maintenance windows for a vendor. Returns full incident timeline — each investigator update, affected components, and resolution. Filter by status to focus on active incidents (use before deploy), resolved history (for postmortem), or maintenance windows: upcoming and in-progress ones, plus completed ones on Atlassian Statuspage, incident.io, Firehydrant, and Datadog vendors (Status.io vendors list upcoming and in-progress only). Narrow further with match (terms in the title or affected component names, plus location names and IDs on Google Cloud and AWS) and since/until (a start-date range) — all applied before paging, so totals count only what they keep. Page through long histories with limit + offset — a truncated result discloses the total and returns the value to page with in nextOffset. Each response is held to 64,000 bytes per surface: a page can stop short of limit at that budget (the paging fields and notice say so), and an incident too large on its own keeps only its newest updates, with updates_omitted counting the rest — full_updates: true returns them all. Some vendor feeds cap their own history: when upstreamCeiling is present the vendor API returned everything it will serve, and older incidents are not reachable at a higher offset; the notice says whether since reaches them. On Atlassian Statuspage vendors, since (a date) also reads the status page's quarterly history archive back to that date, or 24 months back at most, where the page publishes one — not on a vendor that covers only part of a shared page, such as "sendgrid"; those records are marked source: "history" and carry the title, impact, start and end times, and final update message only. On incident.io vendors (such as "openai"), whose feed lists the last 90 days by default, since reads the page's incident list one calendar quarter at a time back to that date, 24 months back at most, and returns full records (source: "api"). An empty result, or a history read that stopped short of since, explains itself in notice.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • devops_list_vendorsList vendors in the built-in registry, optionally filtered by category or name search.No findings

    Why it is in this group

    • The name "devops_list_vendors" says "list": a word for reading.
    Description, as the server wrote it
    List vendors in the built-in registry, optionally filtered by category or name search. Returns slug, display name, category, and status page URL for each entry. Use to discover the correct slug to pass to other tools, or to see which vendors are available before configuring a stack.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: true
  • devops_status_checkCheck the current health status for one or more vendors.Reads text others write1 finding

    Why it is in this group

    • The name "devops_status_check" says "status": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Info

      The input schema at properties.vendors.description mentions "credentials".

      Matched text

      inputSchema.properties.vendors.description: "...se URLs, read at their origin and path (credentials, query, and fragment dropped). Most non..."

      What this check looks for
    Description, as the server wrote it
    Check the current health status for one or more vendors. Accepts registered vendor slugs (e.g., "github", "aws", "gcp", "gitlab") or raw Atlassian Statuspage base URLs. Registry entries are served by each vendor's native status API (Statuspage, incident.io, Status.io, Slack, AWS Health, Google Cloud Service Health, Azure status, Azure DevOps status, Firehydrant, Datadog) and normalized to one shape. Returns per-vendor operational indicator as the vendor published it (none = no degradation reported, minor, major, critical, maintenance = scheduled window), degraded components, and active incidents; summary counts each vendor at the worse of its indicator and its worst open incident rated minor or worse. Use mode: "detailed" for component lists and maintenance windows, narrowed with component_filter, bounded by component_limit, and paged with component_offset. Batch-friendly — pass a list to check your full stack in one call; up to 20 vendors are checked per call, and any past the 20th come back in deferred_vendors to resend. A vendor that cannot be resolved or reached is reported in its own result row, so one bad entry never discards the rest. Every vendor with an active problem gets a pre-filled devops_suggest_action call in nextToolSuggestions.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • devops_suggest_actionReturn an incident-response playbook tailored to a vendor degradation, with pre-filled follow-up tool calls.No findings

    Why it is in this group

    • The description's first verb is "return": a word for reading.
    Description, as the server wrote it
    Return an incident-response playbook tailored to a vendor degradation, with pre-filled follow-up tool calls. Synthesizes category-specific guidance (cloud, CDN, dev-platform, auth, etc.) from built-in incident knowledge and the provided context. Use after devops_status_check or devops_get_incidents surfaces a problem to determine what to investigate next.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false
    • idempotentHint: 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.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or outbound.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write5
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

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.

Tools listed without sign-in

Looked at Nov 20, 2026, 19:29 UTC, unchanged since Oct 22, 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
1
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 1 tool that changes things (1 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://devops-status.caseyjhand.com/mcp: tools/list returned 7 tools to an anonymous client, including devops_watch_stack (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

4 info
  • The input schema at properties.vendors.description mentions "credentials".

    Matched text

    inputSchema.properties.vendors.description: "...se URLs, read at their origin and path (credentials, query, and fragment dropped). Most non..."

    What this check looks for
  • The input schema at properties.vendor.description mentions "credentials".

    Matched text

    inputSchema.properties.vendor.description: "...base URL, read at its origin and path (credentials, query, and fragment dropped). Most oth..."

    What this check looks for
  • The description is 2,060 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...result, or a history read that stopped short of since, explains itself in notice."

    What this check looks for
  • The input schema at properties.vendors.description mentions "credentials".

    Matched text

    inputSchema.properties.vendors.description: "...ead and saved at their origin and path (credentials, query, and fragment dropped). Most oth..."

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. Tools3 changed.~3 changedView diff
  2. Tools2 changed.~2 changedView diff
  3. Tools3 changed.~3 changedView diff

Install

Pinned to version 0.11.2. Values in angle brackets are yours to fill in.

claude mcp add --transport http devops-status https://devops-status.caseyjhand.com/mcp
claude mcp add --transport http devops-status https://devops-status.caseyjhand.com/mcp

Run 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.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__devops-status__devops_list_vendors",
      "mcp__devops-status__devops_status_check",
      "mcp__devops-status__devops_get_incidents",
      "mcp__devops-status__devops_check_certs",
      "mcp__devops-status__devops_check_dns",
      "mcp__devops-status__devops_suggest_action"
    ],
    "ask": [
      "mcp__devops-status__devops_watch_stack"
    ],
    "deny": []
  }
}

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 devops-status. Review it before you use it.

Capability tags behind 5 rules
  • mcp__devops-status__devops_status_check untrusted-input
  • mcp__devops-status__devops_get_incidents untrusted-input
  • mcp__devops-status__devops_watch_stack untrusted-input
  • mcp__devops-status__devops_check_certs untrusted-input
  • mcp__devops-status__devops_check_dns 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/io.github.cyanheads/devops-status-mcp-server/verify?hash=6cd4fa3a33b9e10eb3c4af5fb81ab913e4e51debbd343fcca9a83995096732a5" | jq -e '.changed == false'

Pinned hash 6cd4fa3a33b9, from Nov 16, 2026, 11:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/io.github.cyanheads/devops-status-mcp-server/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for io.github.cyanheads/devops-status-mcp-server
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.cyanheads/devops-status-mcp-server.svg)](https://radar.anlyon.com/servers/io.github.cyanheads/devops-status-mcp-server)

Source: MCP Registry · repository. Synced Nov 16, 2026, 11:17 UTC.

devops-status-mcp-server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the devops-status-mcp-server safe?

No scan can promise that. What its definitions show: 7 tools, 1 of which can change, send or delete something, and 4 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 devops-status-mcp-server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the devops-status-mcp-server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the devops-status-mcp-server have?

7 tools: 1 write, 6 read. devops_watch_stack, devops_list_vendors, devops_status_check, devops_get_incidents, devops_check_certs, devops_check_dns and devops_suggest_action.

Does the devops-status-mcp-server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the devops-status-mcp-server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the devops-status-mcp-server changed its tools?

Yes, 3 times since Anlyon Radar started tracking it, most recently on Nov 16, 2026. Every change is on this page and in its Atom feed.