Skip to content

BitePods MCP server

com.bitepods/bitepodsbitepods.comstreamable-httpv1.0.0Scanned

Free podcast clip maker: share the exact moment of an episode from YouTube, Spotify or Apple.

7 tools: 1 change things and 6 only read.

Tools: 0 delete, 0 irreversible, 1 write, 6 read.
  • Reads private data · 1
  • Reads text others write · 7
  • Can send data out · 4
  • No 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 BitePods MCP server can do

The BitePods MCP server is published by bitepods.com, a domain the MCP Registry verified. It has 7 tools. It can create or change data (1 tool: create_clip), but cannot delete data. 6 tools only read.

1 tool reads private data (search_shows); 7 tools bring in text other people can write (search_shows, top_shows, list_episodes and 4 more); 4 tools can send data out of your system (list_episodes, resolve_url, get_transcript 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.

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.

Last scanned Nov 10, 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 1Claude Code: ask
  • create_clipMarks a span of an episode and hands back the links that play it.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "create_clip" 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
    Marks a span of an episode and hands back the links that play it. Nothing is downloaded, transcoded or re-hosted: a clip is a pointer — the publisher's own URL plus an in and an out point — so the whole clip fits in its link and this call cannot fail for reasons of storage. The episode's name and cover are resolved and returned, so you can check you clipped what you meant to before sharing the link. Times are seconds from the start of the episode, decimals allowed. Returns `{ clip, links: { page, short, sourceAtMoment, mp3 }, shareText }`. Usually followed by: share links.short (or links.page); links.mp3 downloads the audio.

    Annotations the server declares

    • title: Make a shareable clip
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
Only reads 6Show 6 read-only tools
  • get_clipWhat a short id stands for: the episode, the two times, the links, and how many times the clip has been played.Reads text others writeNo findings

    Why it is in this group

    • The name "get_clip" 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.
    Description, as the server wrote it
    What a short id stands for: the episode, the two times, the links, and how many times the clip has been played. The id is the `/x/<id>` part of a short link. Play counts are approximate by design — a per-clip counter with no atomic increment behind it, not an audit trail. Returns `{ clip, links, plays, shareText }`.

    Annotations the server declares

    • title: Read a clip back by its short id
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
  • get_transcriptThe episode's own transcript with absolute timestamps in seconds, so a line's `start` is a number you can pass straight to create_clip.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "get_transcript" 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.
    • 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
    The episode's own transcript with absolute timestamps in seconds, so a line's `start` is a number you can pass straight to create_clip. This is how you find the moment worth clipping: read the transcript, choose the span, clip it. These are the publisher's transcripts (Podcasting 2.0 `<podcast:transcript>`), and only about one show in six ships one. There is no speech-to-text here — deliberately, and it is not coming — so an episode without one answers `available: false` with a reason rather than an approximation. Cue-timed, not word-timed: a long cue is split into readable lines and `estimated: true` says the split points are interpolated between two known times. Returns `{ available, lines, lineCount, truncated, estimated, episode, … }`; `available: false` carries `reason`. Usually followed by: create_clip with a line's start and end.

    Annotations the server declares

    • title: Read the publisher's transcript, timed
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
  • list_episodesEpisodes of a show, newest first, from its RSS feed or its Apple Podcasts page.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "list_episodes" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe description's first sentence says "rss": it reads web pages, which anyone can write.
    • 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
    Episodes of a show, newest first, from its RSS feed or its Apple Podcasts page. Feeds are read up to 60 items — that is the cap, not a page you can walk past. Each episode carries a `clipUrl`: pass it back as `url` to `create_clip` or `get_transcript` and the right episode is clipped, whatever the show's pages look like. Errors with `not_a_show` if the URL is one episode rather than a show. An episode from a feed may also carry `chapters` — the publisher's own timestamp index, read out of the show notes, as `{ startSeconds, title }` in the order they happen. That is a human being naming the parts of their own episode, so it is the cheapest way to find a moment worth clipping: pick a chapter, pass its `startSeconds` to `create_clip` as `start`, and the next chapter's `startSeconds` as `end`. Roughly two shows in five publish one, and a show that doesn't simply has no `chapters` field. Returns `{ show, episodes, total, offset, limit, hasMore }`. An episode may carry `chapters`. Usually followed by: An episode's own `chapters` are the quickest way to a moment; get_transcript reads the words where a show publishes one. Then create_clip.

    Annotations the server declares

    • title: List a show's episodes
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
  • resolve_urlTurns any pasted link into something clippable, and says which of three things it is: `episode` (one playable thing, ready for create_clip), `show` (a feed or channel — a preview o…Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "resolve_url" says "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.
    • untrusted-inputThe name "resolve_url" says "resolve" and "url": it reads web pages, which anyone can write.
    • 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
    Turns any pasted link into something clippable, and says which of three things it is: `episode` (one playable thing, ready for create_clip), `show` (a feed or channel — a preview of its episodes, with list_episodes for the rest), or `unresolved` (nothing playable was found, and why). Answering `unresolved` is a real answer, not a failure: clipping the wrong audio confidently is worse. Spotify links resolve to the same episode in the show's own feed, because nothing else can play Spotify. Returns `{ type: 'episode' | 'show' | 'unresolved', … }` — always 200; read `type`. Usually followed by: create_clip for an episode, list_episodes for a show.

    Annotations the server declares

    • title: Work out what a URL points at
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
  • search_showsSearches the public podcast directory by show name.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "search_shows" says "search": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "search" and "directory": it reads files.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Searches the public podcast directory by show name. Use it when you have a name rather than a link. Each hit carries the show's RSS feed as `sourceUrl` — hand that to `list_episodes`. It searches shows, never episodes: there is no index of episode titles to search, so finding one episode means listing a show's episodes and matching the title yourself. Shows Apple rates explicit are left out. Returns `{ shows }` — id, name, author, artwork, feed, and the `sourceUrl` to list episodes from. Usually followed by: list_episodes with a show's sourceUrl.

    Annotations the server declares

    • title: Find podcasts by name
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
  • top_showsThe most-subscribed shows in a country, ranked.Reads text others writeNo findings

    Why it is in this group

    • The name "top_shows" says "show": 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
    The most-subscribed shows in a country, ranked. Useful as a starting point when nothing specific was named, and as a way to check that a show is the well-known one rather than a similarly named feed. Country defaults to the one the request came from. Returns `{ country, shows }`, each show carrying its 1-based `rank`. Usually followed by: list_episodes with a show's sourceUrl.

    Annotations the server declares

    • title: The podcast charts
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: 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.

28 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 write7
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.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__bitepods__search_shows"
    ]
  }
}

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 18, 2026, 16:20 UTC, unchanged since Nov 11, 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://bitepods.com/mcp: tools/list returned 7 tools to an anonymous client, including create_clip (write)

    What this check looks for

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.

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 bitepods https://bitepods.com/mcp
claude mcp add --transport http bitepods https://bitepods.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__bitepods__search_shows",
      "mcp__bitepods__top_shows",
      "mcp__bitepods__get_clip"
    ],
    "ask": [
      "mcp__bitepods__list_episodes",
      "mcp__bitepods__resolve_url",
      "mcp__bitepods__get_transcript",
      "mcp__bitepods__create_clip"
    ],
    "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 bitepods. Review it before you use it.

Capability tags behind 7 rules
  • mcp__bitepods__search_shows private-data, untrusted-input
  • mcp__bitepods__top_shows untrusted-input
  • mcp__bitepods__list_episodes untrusted-input, outbound
  • mcp__bitepods__resolve_url untrusted-input, outbound
  • mcp__bitepods__get_transcript untrusted-input, outbound
  • mcp__bitepods__create_clip untrusted-input, outbound
  • mcp__bitepods__get_clip 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.bitepods/bitepods/verify?hash=3f9808ca70c4395f43d795878af1f9c6ef22c7108210d5536346539fce2b0ca7" | jq -e '.changed == false'

Pinned hash 3f9808ca70c4, from Nov 10, 2026, 09:18 UTC.

Watch

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

https://radar.anlyon.com/servers/com.bitepods/bitepods/changes.xml

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

Badge

Anlyon Radar badge for com.bitepods/bitepods
[![Anlyon Radar](https://radar.anlyon.com/badge/com.bitepods/bitepods.svg)](https://radar.anlyon.com/servers/com.bitepods/bitepods)

Source: MCP Registry. Synced Nov 10, 2026, 10:17 UTC.

BitePods MCP server: questions and answers

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

Is the BitePods MCP server safe?

No scan can promise that. What its definitions show: 7 tools, 1 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 BitePods MCP server delete data?

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

Can the BitePods 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 BitePods MCP server have?

7 tools: 1 write, 6 read. create_clip, search_shows, top_shows, list_episodes, resolve_url, get_transcript and get_clip.

Does the BitePods MCP server need sign-in or OAuth?

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

Does the BitePods 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): list_episodes, resolve_url, get_transcript and create_clip.

Has the BitePods MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 10, 2026.