Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

Skwirl

REMOTE · SKWIRL.APP

Search saved Instagram, YouTube and X reels and posts as notes from Claude and ChatGPT.

Available components

74 Trust /100

Last scanned (23 hours ago)

Scores higher than 61% of remote servers

Reachable. Last reached 21 Nov 2026.

Category Search and Retrieval

Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. How we score → Why this is hard to score →

Endpoint Security83
  • The endpoint's TLS certificate is valid, in date, and uses a strong key. View diagnostics → Pass
  • Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token. View diagnostics → Pass
  • HTTPS enforcement could not be verified: the plaintext port answered with HTTP 401, which proves neither a plaintext path nor enforcement. View diagnostics → Unverified
  • The HSTS (Strict-Transport-Security) header is present. View diagnostics → Pass
  • DNSSEC check failed: this domain isn't protected by DNSSEC. See how to fix → View diagnostics → Fail
  • The authorisation server offers only Dynamic Client Registration (RFC 7591), which MCP 2026-07-28 deprecated in favour of Client ID Metadata Documents. View diagnostics → Partial
Transport & Reachability100
Schema Quality & AI Usability69
  • AI-judged instruction clarity (excellent).Pass
  • Context-footprint check failed: tool/resource definitions use about 1533 tokens (~219/item across 7 items; 7 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management0
  • Stability check failed: schema churn in the 5 days we've observed: 0 tool removals, 3 breaking changes, 0 auth/transport breaks, 0 additions. See how to fix → Fail
Tool Coverage100
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 100% of tool parameters carry a description.Pass
  • Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • We read all 7 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
  • An AI judge read all 8 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
What we tested

Checks we ran (12)

  • TLS certificate
  • Authorization
  • HSTS header
  • DNSSEC
  • Transport
  • Schema quality
  • Stability
  • Tool coverage
  • Injection markers
  • Destructive annotations
  • Judged manipulation
  • MCP protocol

Checks we could not complete (1)

These are gaps on our side, not findings about this hosted endpoint. We only credit what we can confirm, so a category made up entirely of them scores 0.

  • HTTPS in Endpoint SecurityHTTPS enforcement could not be verified: the plaintext port answered with HTTP 401, which proves neither a plaintext path nor enforcement.

Does not apply to this hosted endpoint (2)

  • Supply Chain SecurityScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.
  • Provenance & TransparencyScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.

The score does not measure everything. What the score does not measure →

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 19 Nov 26 +3
    • Stability: 0.07 → fail ▼ security
    • A breaking change shipped without a version bump: still 0.2.1 ▼ security
    • HSTS header: fail → pass ▲ security
    • Tool “get_stash_note” rewrote its description, which is the text the model reads security
    • Tool “list_stash_mentions” rewrote its description, which is the text the model reads security
    • Tool “list_stash_topics” rewrote its description, which is the text the model reads security
    • Tool “mark_stash_used” rewrote its description, which is the text the model reads security
    • Tool “recent_stash” rewrote its description, which is the text the model reads security
    • Tool “save_link” rewrote its description, which is the text the model reads security
    • Tool “search_stash” rewrote its description, which is the text the model reads security
    • Tool “mark_stash_used” is now declared destructive security
    • Schema quality: pass → fail ▼ functional
    • “list_stash_mentions” changed the type of “limit”: number → integer ▼ functional
    • “recent_stash” changed the type of “limit”: number → integer ▼ functional
    • “search_stash” changed the type of “limit”: number → integer ▼ functional
    • Tool coverage: 60% → 100% ▲ functional
    • Tool “get_stash_note” now declares an output schema ▲ functional
    • Tool “list_stash_mentions” now declares an output schema ▲ functional
    • Tool “list_stash_topics” now declares an output schema ▲ functional
    • Tool “mark_stash_used” now declares an output schema ▲ functional
    • Tool “recent_stash” now declares an output schema ▲ functional
    • Tool “save_link” now declares an output schema ▲ functional
    • Tool “search_stash” now declares an output schema ▲ functional
    • First check of Tool coverage: 100 functional
    • Schema quality: good → excellent functional
    • “get_stash_note” reworded the description of “note_id” cosmetic
    • “list_stash_mentions” reworded the description of “kind” cosmetic
    • “list_stash_mentions” reworded the description of “limit” cosmetic
    • “mark_stash_used” reworded the description of “note_id” cosmetic
    • “recent_stash” reworded the description of “limit” cosmetic
    • “save_link” reworded the description of “note” cosmetic
    • “save_link” reworded the description of “url” cosmetic
    • “search_stash” reworded the description of “limit” cosmetic
    • “search_stash” reworded the description of “query” cosmetic
    • “search_stash” reworded the description of “status” cosmetic
  • 17 Nov 26 +1
    • Stability: unverified → 0.03 ▲ functional
  • 16 Nov 26 70

    First indexed and scored.

Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 21 Nov 2026 · Probed https://skwirl.app/mcp

TLS valid

Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=skwirl.app CN=WE1,O=Google Trust Services,C=US 9 Nov 2026 7 Feb 2027 ECDSA 256 ECDSA-SHA256 39d6fce7dfa733500e5710bda8f708ed
SANs: skwirl.app, www.skwirl.app, *.www.skwirl.app
CN=WE1,O=Google Trust Services,C=US (CA) CN=GTS Root R4,O=Google Trust Services LLC,C=US 13 Dec 2023 20 Feb 2029 ECDSA 256 ECDSA-SHA384 7ff31977972c224a76155d13b6d685e3
CN=GTS Root R4,O=Google Trust Services LLC,C=US (CA) CN=GlobalSign Root CA,OU=Root CA,O=GlobalSign nv-sa,C=BE 15 Nov 2023 28 Jan 2028 ECDSA 384 SHA256-RSA 7fe530bf331343bedd821610493d8a1b

Background: What to check on a remote MCP endpoint →

DNSSEC insecure

Validation of skwirl.app. — Not signed

Zone DS Keys Algorithms Outcome
. Trust anchor 20326, 38696 8, 8 Verified
app. DS published 23684 8 Verified
skwirl.app. No DS Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation
Authentication Enforced and verified

The endpoint asked for a token and published valid RFC 9728 metadata describing how to get one.

Result Enforced and verified
Enforced On tool calls
HTTP status 200

WWW-Authenticate challenge Bearer resource_metadata="https://skwirl.app/.well-known/oauth-protected-resource"

Bearer resource_metadata="https://skwirl.app/.well-known/oauth-protected-resource"
Header Value
strict-transport-security max-age=31536000; includeSubDomains

Protected resource metadata

Document https://skwirl.app/.well-known/oauth-protected-resource
Retrieved Yes
Resource https://skwirl.app/mcp
Authorisation server https://skwirl.app

Background: How OAuth 2.1 works in the 2026 MCP spec →

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://skwirl.app/mcp Verified 200
http (plaintext) http://skwirl.app/mcp Inconclusive 401
MCP tools · 7 exposed · ~1,382 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
get_stash_note ~193

Read one Skwirl note by its exact note_id from search_stash, recent_stash or list_stash_mentions. Returns stored Markdown including summary, next step, transcript and source permalink when available; a missing or inaccessible ID returns a no-match message. Each successful call increments the note's open counter, so repeated reads have an additional side effect. Use search_stash to find relevant notes, recent_stash to browse newest saves, or list_stash_mentions to find named entities before opening one. Usage examples: get_stash_note({"note_id":"a1b2c3d4e5f67890"}). Returns text (also in structuredContent.text).

NameTypeReqDescription
note_idstringyesNonempty string ID returned by search_stash, recent_stash or list_stash_mentions; use it verbatim. Example: a1b2c3d4e5f67890.
NameTypeReqDescription
textstringyesTool result as human-readable text or Markdown, including empty-result and operation status messages.

No examples provided.

list_stash_mentions ~223

List named entities across the connected user's Skwirl saves, rather than the posts' main topics. Read-only; optionally filter kind and set limit from 1–200 (default 50). Returns text entries with name, type, done flag when present, newest source-note title/ID and repeat count when present. Entities are deduplicated by type/name and ordered by newest source note first; ties have no guaranteed order. Returns an empty-result message if none. Use for saved books, places or people; use search_stash for concepts and get_stash_note for source details. Usage examples: list_stash_mentions({"kind":"book","limit":50}); list_stash_mentions({}). Returns text (also in structuredContent.text).

NameTypeReqDescription
kindstring–Optional entity type string, one of book, movie, show, podcast, place, product, person, other; omit for all types. Example: book.
limitinteger–Integer result count from 1 to 200; defaults to 50. Example: 50.
NameTypeReqDescription
textstringyesTool result as human-readable text or Markdown, including empty-result and operation status messages.

No examples provided.

list_stash_topics ~109

List topics in the connected user's Skwirl library with note counts, highest count first; ties have no guaranteed order. Read-only and takes no parameters. Returns text lines in topic: count form, or 'No saves yet.' Use to understand the library's subject mix before search_stash; use list_stash_mentions for named books, places or other entities, or recent_stash for newest notes. Usage examples: list_stash_topics({}). Returns text (also in structuredContent.text).

Input schema present but exposes no named parameters.

NameTypeReqDescription
textstringyesTool result as human-readable text or Markdown, including empty-result and operation status messages.

No examples provided.

mark_stash_used ~208

Record that a Skwirl note shaped the user's work, decision or action; call after applying it, rather than merely reading it. Uses an exact note_id from search_stash, recent_stash or list_stash_mentions. Sets status to used and replaces used_at with the current timestamp; tracks use and changes status-filter results, with no search-ranking boost. Repeating keeps status used but refreshes the timestamp, so it is not fully idempotent. Returns 'Marked as used.' or a no-match message for a missing or inaccessible ID; note content is retained. Usage examples: mark_stash_used({"note_id":"a1b2c3d4e5f67890"}). Returns text (also in structuredContent.text).

NameTypeReqDescription
note_idstringyesNonempty string ID returned by search_stash, recent_stash or list_stash_mentions; use it verbatim. Example: a1b2c3d4e5f67890.
NameTypeReqDescription
textstringyesTool result as human-readable text or Markdown, including empty-result and operation status messages.

No examples provided.

recent_stash ~166

List the connected user's newest Skwirl saves, newest creation date first. Recent means the latest limit notes across all time, with no time window: limit is 1–20, default 5. Read-only; returns compact hits with title, summary (up to 140 characters), note ID, date, topic, status and tools, or 'No matching saves.' Use for browsing recent additions; use search_stash for relevance, list_stash_mentions for named entities, and get_stash_note for a full note. Usage examples: recent_stash({"limit":5}); recent_stash({}). Returns text (also in structuredContent.text).

NameTypeReqDescription
limitinteger–Integer result count from 1 to 20; defaults to 5. Example: 5.
NameTypeReqDescription
textstringyesTool result as human-readable text or Markdown, including empty-result and operation status messages.

No examples provided.

save_link ~231

Queue a supported Instagram, Facebook, YouTube or X link in Skwirl. Requires an authenticated Skwirl connection and the user's intent to save the link; do not use just to read it. Writes a capture for asynchronous processing; success says the note should be ready in a few minutes, not that it is ready now. An existing normalized link in this account returns 'Already saved.' without replacing its reminder. Unsupported links, permission/plan restrictions or queue limits return explanatory failure text. Use search_stash or recent_stash to find a processed note later. Usage examples: save_link({"url":"https://www.instagram.com/reel/ABC123/","note":"Try this recipe"}). Returns text (also in structuredContent.text).

NameTypeReqDescription
notestring–Optional free-text string, up to 500 characters, explaining why the user wants to save this link; omit if not supplied. Example: Try this recipe.
urlstringyesHTTPS URL string for a supported Instagram, Facebook, YouTube or X post. Example: https://www.instagram.com/reel/ABC123/.
NameTypeReqDescription
textstringyesTool result as human-readable text or Markdown, including empty-result and operation status messages.

No examples provided.

search_stash ~252

Search the connected user's Skwirl saves by keyword and meaning before work or planning, or to recover something they saved. Read-only: returns relevance-ordered compact hits with title, summary (up to 140 characters), note ID, date, topic, status and tools, or 'No matching saves.' Use concept words in query; limit is 1–20 (default 5), and optional status is unused or used. Open a relevant ID with get_stash_note; use recent_stash for newest saves or list_stash_mentions for named entities. Usage examples: search_stash({"query":"weeknight pasta","limit":5}); search_stash({"query":"agent memory","status":"unused"}). Returns text (also in structuredContent.text).

NameTypeReqDescription
limitinteger–Integer result count from 1 to 20; defaults to 5. Example: 5.
querystringyesNonempty string of concept words, names or the underlying problem, up to 200 characters. Try a second query if the first is weak. Example: weeknight pasta.
statusstring–Optional exact status string: unused or used; omit for all saves. Example: unused.
NameTypeReqDescription
textstringyesTool result as human-readable text or Markdown, including empty-result and operation status messages.

No examples provided.

Install

How do I install the Skwirl MCP server?

Skwirl is a hosted endpoint at https://skwirl.app/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

remote · skwirl.app

# add to Claude Code
claude mcp add --transport http parthuss-skwirl 'https://skwirl.app/mcp'
Common questions

What is the Skwirl MCP server?

Skwirl is an MCP server listed in the public MCP registry as io.github.Parthuss/skwirl. Search saved Instagram, YouTube and X reels and posts as notes from Claude and ChatGPT. This page covers its hosted endpoint (https://skwirl.app/mcp).

Is the Skwirl MCP server safe to use?

Skwirl scores 74 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the Skwirl MCP server expose?

Skwirl exposes 7 tools: search_stash, get_stash_note, recent_stash, list_stash_topics, list_stash_mentions, and 2 more. Their descriptions and schemas cost roughly 1,382 tokens of context every time the server is loaded.

Does the Skwirl MCP server require authentication?

Yes. Skwirl asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.

Is the Skwirl MCP server still maintained?

Skwirl is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

What did VerifyMCP test on the Skwirl MCP server?

VerifyMCP ran 12 checks on Skwirl, covering TLS certificate, Authorization, HSTS header, DNSSEC and 8 more. One item could not be completed, a gap on our side rather than a finding about Skwirl: HTTPS. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.