txtreel MCP server
com.txtreel/txtreeltxtreel.comstreamable-httpv1.0.0Scanned
Free iMessage, WhatsApp, Instagram DM and Reddit chat videos (1080x1920 MP4) and screenshots.
7 tools: 7 change things.
- Reads private data · 2
- Reads text others write · 2
- Can send data out · 0
- 3 findings
Tools listed without sign-in. Remote posture
With Anlyon
7 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 txtreel MCP server can do
The txtreel MCP server is published by txtreel.com, a domain the MCP Registry verified. It has 7 tools. It can create or change data (7 tools: chat_validate, chat_render_video, chat_screenshot, reddit_render_video and 3 more), but cannot delete data. None of its tools only reads.
2 tools read private data (chat_screenshot and reddit_screenshot); 2 tools bring in text other people can write (chat_screenshot and reddit_screenshot).
The fixed checks matched 3 findings (annotation-conflict and long-description). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 14, 2026, version 1.0.0.
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.
7 tools
Changes things 7Claude Code: ask
chat_render_videoRender a txtreel conversation (script or messages) to an MP4 and wait for it to finish.No findings
Why it is in this group
- The input takes a statement or command to run (input.script), which can change anything it can reach.
Description, as the server wrote itContains invisible characters, shown as [U+…] Render a txtreel conversation (script or messages) to an MP4 and wait for it to finish. This queues a render, waits for it, and returns once it is done (or after failing/timing out). A short render (a handful of messages) typically takes 20-60 seconds. Returns a public link to the MP4; the file is deleted after 24 hours. If txtreel is busy or you hit the rate limit, the result is an error with the message. Conversation script format (one event per line): them: hey, you up? me: yeah why [pause 1.5] wait 1.5 s --- Today 9:41 PM date/time separator me: [photo URL] caption a photo (URL, local file path, or an uploaded photo's number) [them reacts ❤[U+FE0F]] reaction on my last message [read] they read my messages === everything above is already on screen at the start === scroll 3 same, but open on the oldest message and skim down in 3 s # comment "Name: text" also means "them" when Name matches contact.name (e.g. contact.name "Sam" lets you write "Sam: omg" instead of "them: omg"). A literal "\n" inside a message becomes a line break. Lines starting with # are comments. Photos: "me: [photo https://example.com/image.jpg] optional caption" (https URLs only; local file paths are not available). "=== scroll 3" opens on the oldest message of a longer history and skims down to the live part in 3 s. With keyboard true (the default), the newest messages stay above the area where the Reels caption sits.
Annotations the server declares
- title: Render a txtreel video
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
chat_screenshotRender a single frame of a txtreel conversation (script or messages) to a PNG and return it as an image.Reads private dataReads text others writeNo findings
Why it is in this group
- The name "chat_screenshot" says "screenshot": a word for reading.
- The input takes a statement or command to run (input.script), which can change anything it can reach.
What it can reach
- private-dataThe name "chat_screenshot" says "screenshot" and "chat": it reads messages.
- private-dataThe description's first sentence says "conversation": it reads messages.
- untrusted-inputThe name "chat_screenshot" says "screenshot" and "chat": it reads messages, which other people can write.
- untrusted-inputThe description's first sentence says "conversation": it reads messages, which other people can write.
Description, as the server wrote itContains invisible characters, shown as [U+…] Render a single frame of a txtreel conversation (script or messages) to a PNG and return it as an image. Defaults to the last frame — pass "frame" to capture an earlier moment (30 fps). Returns the PNG as image content, plus a text line with the PNG's URL (expires after 24 hours). Conversation script format (one event per line): them: hey, you up? me: yeah why [pause 1.5] wait 1.5 s --- Today 9:41 PM date/time separator me: [photo URL] caption a photo (URL, local file path, or an uploaded photo's number) [them reacts ❤[U+FE0F]] reaction on my last message [read] they read my messages === everything above is already on screen at the start === scroll 3 same, but open on the oldest message and skim down in 3 s # comment "Name: text" also means "them" when Name matches contact.name (e.g. contact.name "Sam" lets you write "Sam: omg" instead of "them: omg"). A literal "\n" inside a message becomes a line break. Lines starting with # are comments. Photos: "me: [photo https://example.com/image.jpg] optional caption" (https URLs only; local file paths are not available). "=== scroll 3" opens on the oldest message of a longer history and skims down to the live part in 3 s. With keyboard true (the default), the newest messages stay above the area where the Reels caption sits.
Annotations the server declares
- title: Screenshot a txtreel conversation
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
chat_validateChecks a txtreel conversation (script or messages) without rendering anything.1 finding
Why it is in this group
- The name "chat_validate" says "validate": a word for reading.
- The input takes a statement or command to run (input.script), which can change anything it can reach.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "chat_validate" 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":false}; The name "chat_validate" says "validate": a word for reading.
Description, as the server wrote itContains invisible characters, shown as [U+…] Checks a txtreel conversation (script or messages) without rendering anything. It is fast and free. On success returns { ok: true, durationSeconds, messageCount, conversation } where "conversation" is the fully-resolved object (all defaults applied). On failure returns { ok: false, errors: string[] } — one message per problem; when the input was a script, errors are prefixed with the offending line number. Videos longer than the server limit are rejected here too. Conversation script format (one event per line): them: hey, you up? me: yeah why [pause 1.5] wait 1.5 s --- Today 9:41 PM date/time separator me: [photo URL] caption a photo (URL, local file path, or an uploaded photo's number) [them reacts ❤[U+FE0F]] reaction on my last message [read] they read my messages === everything above is already on screen at the start === scroll 3 same, but open on the oldest message and skim down in 3 s # comment "Name: text" also means "them" when Name matches contact.name (e.g. contact.name "Sam" lets you write "Sam: omg" instead of "them: omg"). A literal "\n" inside a message becomes a line break. Lines starting with # are comments. Photos: "me: [photo https://example.com/image.jpg] optional caption" (https URLs only; local file paths are not available). "=== scroll 3" opens on the oldest message of a longer history and skims down to the live part in 3 s. With keyboard true (the default), the newest messages stay above the area where the Reels caption sits.
Annotations the server declares
- title: Validate a txtreel conversation
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
lockscreen_render_videoRender an iOS lock screen with notifications to an MP4 and wait for it to finish: a wallpaper, the big clock, and cards from imessage, whatsapp, instagram, chatgpt, claude, reddit,…1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Findings
- Info
The description is 2,220 characters long, over the 2,000 this check looks for.
What this check looks forMatched text
description ends: "...t, the rest counted as "+3 from WhatsApp and Messages"); "list" shows every card."
Description, as the server wrote it Render an iOS lock screen with notifications to an MP4 and wait for it to finish: a wallpaper, the big clock, and cards from imessage, whatsapp, instagram, chatgpt, claude, reddit, mail, phone, facetime, notes, reminders that land one by one at the seconds you set. This queues a render, waits for it, and returns once it is done (or after failing/timing out). A short render (3-5 notifications) typically takes 20-60 seconds. Returns a public link to the MP4; the file is deleted after 24 hours. If txtreel is busy or you hit the rate limit, the result is an error with the message. If the input is invalid, this returns an error result with the validation messages instead of rendering. Lock screen format (an iOS 26 lock screen with notifications): The video starts on a lock screen (wallpaper, big clock, date, padlock). Like the chat videos, the 9:16 video is the iPhone screen cut below the status bar and above the flashlight and camera buttons. Each notification is a Liquid Glass card from one app (imessage, whatsapp, instagram, chatgpt, claude, reddit, mail, phone, facetime, notes, reminders) with a bold title (usually the sender), a body of up to 4 lines, and a see-through time such as "now" or "2m ago". A card lands at its "at" second (seconds from the start of the video), growing out of the bottom of the screen (or out of the card already there) as on iOS 26; at 0 it is already there. While cards are on screen the lower half of the wallpaper darkens, as on the phone. At least one notification is required. The video ends "endHold" seconds (default 2) after the last card has settled. Wallpaper and avatars are https image URLs or data: URIs (local file paths are not available). "wallpaper" also takes one of Apple's iOS 26 wallpapers by name ("sky", "shadow", "halo", "dusk", "yellow", "pink", "azure", "purple"); without one, "sky" (the iOS 26 default) is drawn, in its light or dark version with "theme". "avatar" on a notification shows the contact's photo with the app icon as a small badge. "display": "stack" folds the cards like iOS 26 does (the newest card, the next one or two as one-line rows under it, the rest counted as "+3 from WhatsApp and Messages"); "list" shows every card.
Annotations the server declares
- title: Render a lock screen notification video
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
lockscreen_screenshotRender a single frame of an iOS lock screen with notifications to a PNG and return it as an image.1 finding
Why it is in this group
- The name "lockscreen_screenshot" says "screenshot": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
Findings
- Warn
The server marks "lockscreen_screenshot" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":false,"openWorldHint":false}; The name "lockscreen_screenshot" says "screenshot": a word for reading.
Description, as the server wrote it Render a single frame of an iOS lock screen with notifications to a PNG and return it as an image. Defaults to the last frame — pass "frame" to capture an earlier moment (30 fps). Returns the PNG as image content, plus a text line with the PNG's URL (expires after 24 hours). If the input is invalid, this returns an error result with the validation messages instead of an image. Lock screen format (an iOS 26 lock screen with notifications): The video starts on a lock screen (wallpaper, big clock, date, padlock). Like the chat videos, the 9:16 video is the iPhone screen cut below the status bar and above the flashlight and camera buttons. Each notification is a Liquid Glass card from one app (imessage, whatsapp, instagram, chatgpt, claude, reddit, mail, phone, facetime, notes, reminders) with a bold title (usually the sender), a body of up to 4 lines, and a see-through time such as "now" or "2m ago". A card lands at its "at" second (seconds from the start of the video), growing out of the bottom of the screen (or out of the card already there) as on iOS 26; at 0 it is already there. While cards are on screen the lower half of the wallpaper darkens, as on the phone. At least one notification is required. The video ends "endHold" seconds (default 2) after the last card has settled. Wallpaper and avatars are https image URLs or data: URIs (local file paths are not available). "wallpaper" also takes one of Apple's iOS 26 wallpapers by name ("sky", "shadow", "halo", "dusk", "yellow", "pink", "azure", "purple"); without one, "sky" (the iOS 26 default) is drawn, in its light or dark version with "theme". "avatar" on a notification shows the contact's photo with the app icon as a small badge. "display": "stack" folds the cards like iOS 26 does (the newest card, the next one or two as one-line rows under it, the rest counted as "+3 from WhatsApp and Messages"); "list" shows every card.
Annotations the server declares
- title: Screenshot a lock screen with notifications
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
reddit_render_videoRender a Reddit thread (script, or post+comments) to an MP4 and wait for it to finish.No findings
Why it is in this group
- The input takes a statement or command to run (input.script), which can change anything it can reach.
Description, as the server wrote it Render a Reddit thread (script, or post+comments) to an MP4 and wait for it to finish. This queues a render, waits for it, and returns once it is done (or after failing/timing out). A short render typically takes 20-60 seconds. Returns a public link to the MP4; the file is deleted after 24 hours. If txtreel is busy or you hit the rate limit, the result is an error with the message. If the thread is invalid, this returns an error result with the validation messages (script errors are prefixed with the offending line number) instead of rendering. Reddit thread script format (post + nested comments): title: AITA for not going to my sister's wedding? body: First paragraph of the post (optional, one line per paragraph) body: Second paragraph [photo URL] a photo in the post (URL, file path or uploaded photo's number) sleepy_joe: A top-level comment > OP: A reply by the post's author (blue OP label) >> sleepy_joe [2.1K 3h]: A reply to that reply, with its votes and age [91 more replies] collapsed replies under the comment above [pause 2] wait 2 s # comment The video is a screen recording of a thread on mobile reddit.com in iOS Safari: the page is laid out once and the view scrolls from block to block, holding each one long enough to read. The title is on screen from frame 0. Scores and ages left out of the script are generated (decreasing down the thread, never older than the post).
Annotations the server declares
- title: Render a Reddit thread video
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
reddit_screenshotRender a single frame of a Reddit thread (script, or post+comments) to a PNG and return it as an image.Reads private dataReads text others writeNo findings
Why it is in this group
- The name "reddit_screenshot" says "screenshot": a word for reading.
- The input takes a statement or command to run (input.script), which can change anything it can reach.
What it can reach
- private-dataThe description's first sentence says "thread": it reads messages.
- untrusted-inputThe description's first sentence says "thread": it reads messages, which other people can write.
Description, as the server wrote it Render a single frame of a Reddit thread (script, or post+comments) to a PNG and return it as an image. Defaults to the last frame — pass "frame" to capture an earlier moment (30 fps). Returns the PNG as image content, plus a text line with the PNG's URL (expires after 24 hours). If the thread is invalid, this returns an error result with the validation messages instead of an image. Reddit thread script format (post + nested comments): title: AITA for not going to my sister's wedding? body: First paragraph of the post (optional, one line per paragraph) body: Second paragraph [photo URL] a photo in the post (URL, file path or uploaded photo's number) sleepy_joe: A top-level comment > OP: A reply by the post's author (blue OP label) >> sleepy_joe [2.1K 3h]: A reply to that reply, with its votes and age [91 more replies] collapsed replies under the comment above [pause 2] wait 2 s # comment The video is a screen recording of a thread on mobile reddit.com in iOS Safari: the page is laid out once and the view scrolls from block to block, holding each one long enough to read. The title is on screen from frame 0. Scores and ages left out of the script are generated (decreasing down the thread, never older than the post).
Annotations the server declares
- title: Screenshot a Reddit thread
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- 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.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged outbound.
- Reads private data2
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write2
- 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.
Looked at Nov 20, 2026, 18:49 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
- 7
- 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 7 tools that change things (7 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://txtreel.com/mcp: tools/list returned 7 tools to an anonymous client, including chat_validate (write), chat_render_video (write), chat_screenshot (write), reddit_render_video (write), reddit_screenshot (write) and 2 more
Findings
Matches of fixed checks, with the text they matched.
- Warnchat_validate
The server marks "chat_validate" 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":false}; The name "chat_validate" says "validate": a word for reading.
The server marks "lockscreen_screenshot" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":false,"openWorldHint":false}; The name "lockscreen_screenshot" says "screenshot": a word for reading.
The description is 2,220 characters long, over the 2,000 this check looks for.
What this check looks forMatched text
description ends: "...t, the rest counted as "+3 from WhatsApp and Messages"); "list" shows every card."
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 1.0.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http txtreel https://txtreel.com/mcpclaude mcp add --transport http txtreel https://txtreel.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": [],
"ask": [
"mcp__txtreel__chat_validate",
"mcp__txtreel__chat_render_video",
"mcp__txtreel__chat_screenshot",
"mcp__txtreel__reddit_render_video",
"mcp__txtreel__reddit_screenshot",
"mcp__txtreel__lockscreen_render_video",
"mcp__txtreel__lockscreen_screenshot"
],
"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 txtreel. Review it before you use it.
Capability tags behind 2 rules
- mcp__txtreel__chat_screenshot private-data, untrusted-input
- mcp__txtreel__reddit_screenshot private-data, 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.txtreel/txtreel/verify?hash=3c774bc8eae7d3b5fdb454ac80e5654711c1d31133f443f01c5ef9ea7b6b7084" | jq -e '.changed == false'Pinned hash 3c774bc8eae7, from Nov 14, 2026, 16:17 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.txtreel/txtreel/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.txtreel/txtreel)Source: MCP Registry. Synced Oct 21, 2026, 18:10 UTC.
txtreel MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the txtreel MCP server safe?
- No scan can promise that. What its definitions show: 7 tools, 7 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 txtreel MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the txtreel 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 txtreel MCP server have?
- 7 tools: 7 write. chat_validate, chat_render_video, chat_screenshot, reddit_render_video, reddit_screenshot, lockscreen_render_video and lockscreen_screenshot.
Does the txtreel MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the txtreel MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the txtreel MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 14, 2026.