OperStack MCP server
com.oper-stack/operstackoper-stack.comstreamable-httpv1.2.0Scanned
Scores whether AI assistants can read, understand and quote a public website. No account.
11 tools: 10 change things and 1 only read.
- Reads private data · 1
- Reads text others write · 5
- Can send data out · 4
- 1 finding
Tools listed without sign-in. Remote posture
With Anlyon
10 tools here change things. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
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 OperStack MCP server can do
The OperStack MCP server is published by oper-stack.com, a domain the MCP Registry verified. It has 11 tools. It can create or change data (10 tools: audit_site, compare_sites, ai_visits, check_counter_installed and 6 more), but cannot delete data. 1 tool only reads.
1 tool reads private data (check_llms_txt); 5 tools bring in text other people can write (audit_site, compare_sites, check_llms_txt and 2 more); 4 tools can send data out of your system (audit_site, compare_sites, check_llms_txt and 1 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 1 finding (annotation-conflict). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 14, 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.
11 tools
Changes things 10Claude Code: ask
ai_visitsReports how many visits each AI assistant sent to a site that runs the free OperStack counter.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Reports how many visits each AI assistant sent to a site that runs the free OperStack counter. Needs the private token from the counter link, which only the site owner has. Returns visit counts per assistant, not individual visitors.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
audit_siteScores any public site out of 100 on the five things an answer engine needs before it will quote a page: whether AI crawlers are allowed in, whether there is a map for agents (llms…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 Scores any public site out of 100 on the five things an answer engine needs before it will quote a page: whether AI crawlers are allowed in, whether there is a map for agents (llms.txt), whether the entity is clear from schema, whether there is anything quotable, and whether pages carry dates and sources. Use it when someone asks whether ChatGPT or another assistant can read and quote a site. Reads up to 10 public pages. A site that turns away automated readers cannot be scored.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
- idempotentHint: true
check_counter_installedLooks at a site page and says whether the OperStack visit counter line is really there, including the case where it is there carrying a different key.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.
- 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.
Description, as the server wrote it Looks at a site page and says whether the OperStack visit counter line is really there, including the case where it is there carrying a different key. Use it when a counter reports nothing and the owner does not know why.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
- idempotentHint: true
check_rivalOperStack Watch subscribers, with their key.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 OperStack Watch subscribers, with their key. Scores a competitor site out of 100 on the same rules as the OperStack free check and the Watch email, and puts it next to the subscriber's own score from the last weekly check.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
- idempotentHint: true
compare_sitesRuns the same AI-visibility scoring on two to four sites and returns them side by side, so a score means something.Reads text others writeCan send data out1 finding
Why it is in this group
- The name "compare_sites" says "compare": a word for reading.
- The description's first verb is "run": a word for a change.
- 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.
- untrusted-inputThe input takes a URL (input.urls), and the tool reads it: it reads web pages, which anyone can write.
- outboundThe input takes a URL (input.urls): a request to a URL can carry data to whoever runs it.
Findings
- Warn
The server marks "compare_sites" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "compare_sites" says "compare": a word for reading.
Description, as the server wrote it Runs the same AI-visibility scoring on two to four sites and returns them side by side, so a score means something. Use it to see where a site is losing answers to a rival, area by area. A site that turns away automated readers is reported as unreadable.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
- idempotentHint: true
fix_of_the_weekOperStack Watch subscribers, with their key.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it OperStack Watch subscribers, with their key. The fix of the week from the last Watch email: the buyer question it answers, who ChatGPT named instead, what to change, and the text to put on the site, word for word as in the email.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
my_questionsOperStack Watch subscribers, with their key.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 OperStack Watch subscribers, with their key. Shows the 10 buyer questions Watch asks ChatGPT and Perplexity every week. Can also replace one question (number and new text) or all ten. Only change questions when the person asked for it. Changes apply from the next weekly check, and history starts again for a changed question.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
my_visibilityOperStack Watch subscribers, with their key.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it OperStack Watch subscribers, with their key. Whether ChatGPT (and Perplexity, when it answered) named the subscriber in answers to their 10 buyer questions at the last weekly check, the trend week by week, and the share of answers against the competitors named. The same numbers as that week's Watch email.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
what_changedOperStack Watch subscribers, with their key.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it OperStack Watch subscribers, with their key. What moved between the last two weekly checks: how many answers named the subscriber in ChatGPT and Perplexity, the share of answers against each competitor, and the buyer questions where they were named or dropped.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
who_insteadOperStack Watch subscribers, with their key.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it OperStack Watch subscribers, with their key. For each of the 10 buyer questions at the last weekly check: whether ChatGPT (and Perplexity) named the subscriber, and which businesses they named instead. Then the sites these answers were built from and where to get mentioned.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: false
- idempotentHint: true
Only reads 1Show 1 read-only tool
check_llms_txtReads a site llms.txt, the file that tells an assistant which pages matter, and follows its links to see whether they still lead anywhere.Reads private dataReads text others writeCan send data outNo findings
Why it is in this group
- The description's first verb is "read": a word for reading.
What it can reach
- private-dataThe description's first sentence says "read" and "file": it reads files.
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Claude Code: ask, because it is tagged outbound.
Description, as the server wrote it Reads a site llms.txt, the file that tells an assistant which pages matter, and follows its links to see whether they still lead anywhere. Use it when someone asks about llms.txt or whether their map for AI assistants works.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- openWorldHint: true
- 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.
20 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 private-data tool breaks every one.
- Reads private data1
- 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 out4
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__operstack__check_llms_txt"
]
}
}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 20, 2026, 18:44 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
- 10
- 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 10 tools that change things (10 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://oper-stack.com/api/mcp/: tools/list returned 11 tools to an anonymous client, including audit_site (write), compare_sites (write), ai_visits (write), check_counter_installed (write), my_visibility (write) and 5 more
Findings
Matches of fixed checks, with the text they matched.
- Warncompare_sites
The server marks "compare_sites" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "compare_sites" says "compare": a word for reading.
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 1.2.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http operstack https://oper-stack.com/api/mcp/claude mcp add --transport http operstack https://oper-stack.com/api/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.
{
"permissions": {
"allow": [],
"ask": [
"mcp__operstack__audit_site",
"mcp__operstack__compare_sites",
"mcp__operstack__check_llms_txt",
"mcp__operstack__ai_visits",
"mcp__operstack__check_counter_installed",
"mcp__operstack__my_visibility",
"mcp__operstack__who_instead",
"mcp__operstack__fix_of_the_week",
"mcp__operstack__what_changed",
"mcp__operstack__check_rival",
"mcp__operstack__my_questions"
],
"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 operstack. Review it before you use it.
Capability tags behind 5 rules
- mcp__operstack__audit_site untrusted-input, outbound
- mcp__operstack__compare_sites untrusted-input, outbound
- mcp__operstack__check_llms_txt private-data, untrusted-input, outbound
- mcp__operstack__check_counter_installed untrusted-input
- mcp__operstack__check_rival untrusted-input, outbound
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.oper-stack/operstack/verify?hash=06098d4c77821c19ae2dd1d2044b5f0d3556b3483c8231391547b5670d2351c4" | jq -e '.changed == false'Pinned hash 06098d4c7782, from Nov 14, 2026, 00:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.oper-stack/operstack/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.oper-stack/operstack)Source: MCP Registry. Synced Oct 21, 2026, 17:59 UTC.
OperStack MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the OperStack MCP server safe?
- No scan can promise that. What its definitions show: 11 tools, 10 of which can change, send or delete something, and 1 finding from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the OperStack MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the OperStack 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 OperStack MCP server have?
- 11 tools: 10 write, 1 read. audit_site, compare_sites, ai_visits, check_counter_installed, my_visibility, who_instead, fix_of_the_week, what_changed and 3 more.
Does the OperStack MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the OperStack MCP server send data to outside services?
- 4 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): audit_site, compare_sites, check_llms_txt and check_rival.
Has the OperStack MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 14, 2026.