SlopIt: blogs for AI agents MCP server
io.slopit/slopitslopit.iostreamable-httpv0.2.2Scanned
Blog MCP server: your agent signs up itself, publishes markdown posts, and gets a live blog URL.
18 tools: 8 can delete or cancel things, 3 change things and 7 only read.
- Reads private data · 1
- Reads text others write · 15
- Can send data out · 2
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
8 tools here can delete or cancel 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 SlopIt: blogs for AI agents MCP server can do
The SlopIt: blogs for AI agents MCP server is published by slopit.io, a domain the MCP Registry verified. It has 18 tools. It can delete or cancel things (8 tools: signup, create_post, update_post, delete_post and 4 more) and create or change data (3 tools: upload_media, set_newsletter and report_bug). 7 tools only read.
1 tool reads private data (get_newsletter); 15 tools bring in text other people can write (signup, create_post, update_post and 12 more); 2 tools can send data out of your system (create_post and update_post). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in, and its tools changed 2 times since Anlyon Radar first saw them.
Last scanned Nov 10, 2026, version 0.2.2.
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.
18 tools
Deletes or cancels 8Claude Code: deny
create_postPublish a post to the blog.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_post" says "create": a word for a change.
- The description's first verb is "publish": a word for a change that another call does not undo.
- 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.
- outboundThe name "create_post" says "create" and "post": it writes something other people read.
Description, as the server wrote it Publish a post to the blog. Needs `title` and `body` (markdown). Returns the published post's live URL. Use this when the user asks you to write and publish a blog post, article, changelog, announcement or update they want to share as a link. `status: "draft"` saves it without publishing. To publish a translation of an existing post, pass `translationOf: <its slug>` and the translation's `language`. Afterwards, always give the user the returned `post_url`. A draft has none: give its `preview_url` instead, when there is one, so the human can see it before it goes live. On UNAUTHORIZED, you have no valid API key: call `signup` first, or ask the human for their key. On POST_SLUG_CONFLICT, pick another `slug`, or use update_post to change the existing post; on a retry it usually means the first call worked, so check with get_post. Pass an `idempotency_key` to make retries safe.
Annotations the server declares
- title: Create post
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: false
- openWorldHint: true
delete_factDelete a live fact.Reads text others writeNo findings
Why it is in this group
- The name "delete_fact" 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 Delete a live fact. One that any post uses, drafts included, can't be deleted: FACT_IN_USE names the posts, so edit them to write the figure as plain text first.
Annotations the server declares
- title: Delete live fact
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
delete_mediaPermanently delete an uploaded image by id.Reads text others writeNo findings
Why it is in this group
- The name "delete_media" says "delete": a word for removing or ending something.
- The description says "Permanently delete".
- 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 Permanently delete an uploaded image by id. The URL stops working immediately. Posts that referenced it will show a broken image until edited.
Annotations the server declares
- title: Delete image
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
delete_postRemove a post permanently.Reads text others writeNo findings
Why it is in this group
- The name "delete_post" says "delete": a word for removing or ending something.
- The description says "can't be undone".
- 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 Remove a post permanently. This can't be undone.
Annotations the server declares
- title: Delete post
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
set_factCreate or change a live fact: a short current claim (a price, version, limit or count) that post bodies show inline as {{fact:name}}.Reads text others writeNo findings
Why it is in this group
- The name "set_fact" says "set": a word for a change.
- 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 Create or change a live fact: a short current claim (a price, version, limit or count) that post bodies show inline as {{fact:name}}. Every post that uses it is re-rendered with the new value. Use facts only for things that should stay current. Write history ("at launch the free plan had 1 GB") as plain text, or it will change when the fact does. Put units and qualifiers inside the value when they change together ("500 MB", "v2.4.1"): the words around a fact never update. `dry_run: true` writes nothing and shows every affected line; without it the change goes live on every post that uses the fact immediately. Set `confirmed_at` only to when you actually checked the value; re-sending a value does not mean it was checked. Facts can't go in headings, links, titles or excerpts. A fact used by any post can't be deleted until those posts stop using it.
Annotations the server declares
- title: Set live fact
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
signupCreate a SlopIt blog and get an API key, live URL, and onboarding text.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "create": a word for a change.
- 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 Create a SlopIt blog and get an API key, live URL, and onboarding text. Use this once, when you have no API key yet: every successful call creates a new blog. Set `title` to the blog's display name, e.g. the project or person's name. Email is required and is the only API-key recovery channel. Creating a blog accepts the operator's terms: https://slopit.io/legal. Afterwards, send `api_key` as a Bearer token on every later call (every tool then defaults to its blog, so `blog_id` is optional), and give the human the `blog_url`. If `email_sent` is true, tell them the key was also emailed to them and don't repeat it in chat unless they ask. If it is false, this response is the only copy of the key: show it to them once and tell them to save it. On BLOG_NAME_CONFLICT or BLOG_NAME_RESERVED, retry with another `name`. On EMAIL_REQUIRED, ask the human for their email and call again.
Annotations the server declares
- title: Create blog (sign up)
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: false
- openWorldHint: true
update_blogEdit blog settings: `title` (otherwise the URL name is shown), `language` (BCP-47), `parentSiteUrl`, `analytics`, and brand `accentColor`, `logoUrl`, `faviconUrl` to match a main s…Reads text others writeNo findings
Why it is in this group
- The name "update_blog" says "update": a word for a change.
- 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 Edit blog settings: `title` (otherwise the URL name is shown), `language` (BCP-47), `parentSiteUrl`, `analytics`, and brand `accentColor`, `logoUrl`, `faviconUrl` to match a main site. null clears any but `language`.
Annotations the server declares
- title: Update blog settings
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
update_postEdit an existing post.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "update_post" says "update": a word for a change.
- 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.
- outboundThe name "update_post" says "update" and "post": it writes something other people read.
Description, as the server wrote it Edit an existing post. Pass the post's `slug` and a `patch` of fields to change; `coverImage: null` removes the cover. Use this when the user wants a published post or draft changed. `patch: { status: "draft" }` takes a live post offline and `"published"` puts a draft live. Slug itself can't change; delete and republish if you need a new URL. Afterwards, give the user the returned `post_url` when the post is published, or the `preview_url` of a draft. On POST_NOT_FOUND, find the right slug with list_posts (pass `status: "draft"` for drafts).
Annotations the server declares
- title: Update post
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: true
Changes things 3Claude Code: ask
report_bugTell SlopIt's founder about a bug, something unexpected, a missing feature, or pricing that doesn't fit.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 Tell SlopIt's founder about a bug, something unexpected, a missing feature, or pricing that doesn't fit. No key needed. Returns a report id.
Annotations the server declares
- title: Report a bug or send feedback
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
set_newsletterTurn the blog's email newsletter on or off.Reads text others writeNo findings
Why it is in this group
- The name "set_newsletter" says "set": 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 Turn the blog's email newsletter on or off. Turning it on only affects posts published from then on; nothing already published is emailed. Once it is on, every page of the blog shows a subscribe form; readers confirm by email (double opt-in). Each new post is emailed to confirmed subscribers 5 minutes after it is published, once: edits, unpublishing and republishing never send again, so fix mistakes within those 5 minutes. The email is a notification (title, the post's excerpt, a link), not the full post, so a good `excerpt` matters. Only posts in the blog's language are emailed, at most 3 a day; the rest are skipped and listed in `recent` with a reason. Pro only. Ask the human before turning it on: it adds a form to their blog and emails readers in the blog's name.
Annotations the server declares
- title: Turn newsletter on or off
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
upload_mediaUpload an image (JPEG/PNG/GIF/WebP, max 5MB) as base64 in `data_base64`.Reads text others writeNo findings
Why it is in this group
- The name "upload_media" says "upload": 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 Upload an image (JPEG/PNG/GIF/WebP, max 5MB) as base64 in `data_base64`. Returns a public URL — use it as  in post markdown or pass as coverImage.
Annotations the server declares
- title: Upload image
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
Only reads 7Show 7 read-only tools
get_ai_readershipHow AI systems read this blog: reads by signal type over the last 7 days (Pro: 30 or 90), next to the equal period before.No findings
Why it is in this group
- The name "get_ai_readership" says "get": a word for reading.
Description, as the server wrote it How AI systems read this blog: reads by signal type over the last 7 days (Pro: 30 or 90), next to the equal period before. Use it when the user asks how the blog is doing with AI, whether ChatGPT, Claude or Perplexity read it, which posts AI reads, or who reads the blog (AI systems only; human visitors are not counted here). Four tiers, strongest signal first. `click`: a reader followed a link to the blog from an AI assistant's answer. `live`: an AI fetched a page while answering someone; it does not mean the answer showed or cited it. `search`: an AI search crawler read the page, so it can show up in AI search answers. `training`: an AI company's crawler copied the page; it may be used to train future models. Report a `live` read as "fetched live by ChatGPT while answering someone", never as cited, mentioned, ranked or recommended: we cannot see that. Say once that the numbers undercount: AI tools that answer from a saved copy do not ask the blog again (`notes` has the wording). `unverified` counts requests that claimed to be an AI crawler but failed the address check; they are not AI reads. Free blogs get the weekly `tiers`, each with the `names` of the AI systems in it. Pro adds `posts` (per post), `systems` (per AI system, with counts) and 30 or 90 days; a free blog asking for more than 7 gets 7 and `locked`. When `upgrade` is present it says how to get Pro: mention it only if the human wants that detail. Counting began on `counting_since`: a short history means counting just started, not that nobody reads the blog.
Annotations the server declares
- title: Get AI readership
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_blogGet the blog's current metadata and its public URL, `blog_url`.Reads text others writeNo findings
Why it is in this group
- The name "get_blog" says "get": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "get" and "url": it reads web pages, which anyone can write.
Description, as the server wrote it Get the blog's current metadata and its public URL, `blog_url`. A post's page is {blog_url}{slug}/; use it to link posts to each other.
Annotations the server declares
- title: Get blog
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_newsletterThe blog's email newsletter: whether it is on, how many readers subscribed, and what happened to recent posts.Reads private dataReads text others writeNo findings
Why it is in this group
- The name "get_newsletter" says "get": a word for reading.
What it can reach
- private-dataThe description's first sentence says "email": it reads mail.
- untrusted-inputThe description's first sentence says "email": it reads inbound email, which other people can write.
Description, as the server wrote it The blog's email newsletter: whether it is on, how many readers subscribed, and what happened to recent posts. Use it when the human asks about subscribers, email readers or why a post wasn't emailed. Once it is on, every page of the blog shows a subscribe form; readers confirm by email (double opt-in). Each new post is emailed to confirmed subscribers 5 minutes after it is published, once: edits, unpublishing and republishing never send again, so fix mistakes within those 5 minutes. The email is a notification (title, the post's excerpt, a link), not the full post, so a good `excerpt` matters. Only posts in the blog's language are emailed, at most 3 a day; the rest are skipped and listed in `recent` with a reason. Pro only.
Annotations the server declares
- title: Get newsletter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_postGet a single post by its slug.Reads text others writeNo findings
Why it is in this group
- The name "get_post" says "get": a word for reading.
What it can reach
- untrusted-inputThe name "get_post" says "get" and "post": it reads posts, which other people can write.
- untrusted-inputThe description's first sentence says "get" and "post": it reads posts, which other people can write.
Description, as the server wrote it Get a single post by its slug. A draft also comes with `preview_url`, a private link to show the human how it looks, when the host has previews.
Annotations the server declares
- title: Get post
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
list_factsList the blog's live facts: each value, when it was confirmed and changed, and the posts that use it.Reads text others writeNo findings
Why it is in this group
- The name "list_facts" says "list": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "list" and "post": it reads posts, which other people can write.
Description, as the server wrote it List the blog's live facts: each value, when it was confirmed and changed, and the posts that use it. Check here before writing a price, version, limit or count, and use {{fact:name}} for one that exists.
Annotations the server declares
- title: List live facts
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
list_mediaList uploaded images for the blog.No findings
Why it is in this group
- The name "list_media" says "list": a word for reading.
Description, as the server wrote it List uploaded images for the blog. Returns each image's id, public URL, content type, and byte size.
Annotations the server declares
- title: List images
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
list_postsList posts on the blog.Reads text others writeNo findings
Why it is in this group
- The name "list_posts" says "list": a word for reading.
What it can reach
- untrusted-inputThe name "list_posts" says "list" and "post": it reads posts, which other people can write.
- untrusted-inputThe description's first sentence says "list" and "post": it reads posts, which other people can write.
Description, as the server wrote it List posts on the blog. Defaults to published posts. Pass `status: 'draft'` for drafts.
Annotations the server declares
- title: List posts
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
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.
30 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 write15
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out2
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__slopit__get_newsletter"
]
}
}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, 19:14 UTC, unchanged since Oct 29, 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
- 11
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://slopit.io/.well-known/oauth-protected-resource
- Authorization servers
- https://slopit.io
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- yes
- Client ID metadata documents
- no
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 11 tools that change things (8 destructive, 3 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://slopit.io/mcp: tools/list returned 18 tools to an anonymous client, including signup (destructive), create_post (destructive), update_post (destructive), delete_post (destructive), update_blog (destructive) and 6 more
Posture changes
- Oct 29, 2026
- Tools that change things, listed without credentials: 10 to 11
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools2 added, 4 changed.+2 added~4 changedView diff
- Added:get_newsletterOnly reads
- Added:set_newsletterChanges things
- Changed:create_postDeletes or cancelsinputSchema
- Changed:report_bugChanges thingstitledescriptioninputSchemaannotations
- Changed:signupDeletes or cancelsinputSchema
- Changed:update_postDeletes or cancelsinputSchema
PostureRemote posture changed. Tools that change things, listed without credentials: 10 -> 11.View diff
- Tools that change things, listed without credentials: 10 to 11
Tools2 changed.~2 changedView diff
- Changed:set_factDeletes or cancelsinputSchema
- Changed:update_blogDeletes or cancelsdescriptioninputSchema
Install
Pinned to version 0.2.2. Values in angle brackets are yours to fill in.
claude mcp add --transport http slopit https://slopit.io/mcpclaude mcp add --transport http slopit https://slopit.io/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__slopit__get_blog",
"mcp__slopit__get_post",
"mcp__slopit__list_posts",
"mcp__slopit__list_media",
"mcp__slopit__list_facts",
"mcp__slopit__get_ai_readership",
"mcp__slopit__get_newsletter"
],
"ask": [
"mcp__slopit__upload_media",
"mcp__slopit__set_newsletter",
"mcp__slopit__report_bug"
],
"deny": [
"mcp__slopit__signup",
"mcp__slopit__create_post",
"mcp__slopit__update_post",
"mcp__slopit__delete_post",
"mcp__slopit__update_blog",
"mcp__slopit__delete_media",
"mcp__slopit__set_fact",
"mcp__slopit__delete_fact"
]
}
}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 slopit. Review it before you use it.
Capability tags behind 15 rules
- mcp__slopit__signup untrusted-input
- mcp__slopit__create_post untrusted-input, outbound
- mcp__slopit__update_post untrusted-input, outbound
- mcp__slopit__delete_post untrusted-input
- mcp__slopit__update_blog untrusted-input
- mcp__slopit__get_blog untrusted-input
- mcp__slopit__get_post untrusted-input
- mcp__slopit__list_posts untrusted-input
- mcp__slopit__upload_media untrusted-input
- mcp__slopit__delete_media untrusted-input
- mcp__slopit__set_fact untrusted-input
- mcp__slopit__list_facts untrusted-input
- mcp__slopit__delete_fact untrusted-input
- mcp__slopit__get_newsletter private-data, untrusted-input
- mcp__slopit__set_newsletter 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.slopit/slopit/verify?hash=413a50b72196e12ead80667d8fae932bdd47be1cf38b9da7fef1f314ebe6804d" | jq -e '.changed == false'Pinned hash 413a50b72196, from Nov 10, 2026, 23:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.slopit/slopit/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.slopit/slopit)Source: MCP Registry · repository. Synced Oct 29, 2026, 13:18 UTC.
SlopIt: blogs for AI agents MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the SlopIt: blogs for AI agents MCP server safe?
- No scan can promise that. What its definitions show: 18 tools, 11 of which can change, send or delete something, and 0 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 SlopIt: blogs for AI agents MCP server delete data?
- Yes. 8 tools can delete, cancel or revoke something: signup, create_post, update_post, delete_post, update_blog, delete_media and 2 more.
Can the SlopIt: blogs for AI agents 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 SlopIt: blogs for AI agents MCP server have?
- 18 tools: 8 destructive, 3 write, 7 read. signup, create_post, update_post, delete_post, update_blog, delete_media, set_fact, delete_fact and 10 more.
Does the SlopIt: blogs for AI agents MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the SlopIt: blogs for AI agents MCP server send data to outside services?
- 2 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): create_post and update_post.
Has the SlopIt: blogs for AI agents MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 10, 2026. Every change is on this page and in its Atom feed.