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.

Mark

REMOTE · MARK.TATASTU.DEV

Mark: stamp files with a signed, timestamped provenance record. Hash locally; verify free.

Available components

57 Trust /100

Last scanned (20 hours ago)

Scores higher than 45% of remote servers

Reachable. Last reached 18 Oct 2026.

Category Security

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 Security46
Transport & Reachability100
Schema Quality & AI Usability68
  • AI-judged instruction clarity (good).Pass
  • Context-footprint check failed: tool/resource definitions use about 1069 tokens (~152/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 Management3
  • Stability observed for 1 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage98
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 94% of tool parameters carry a description.Partial
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.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
Capabilities60
  • Spec-recency check failed: implements MCP spec 2025-06-18; the latest is 2026-07-28. See how to fix → Fail
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.

  • 18 Oct 26 0
    • Stability: unverified → 0.03 ▲ functional
  • 17 Oct 26 57

    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 18 Oct 2026 · Probed https://mark.tatastu.dev/mcp

TLS valid

Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=tatastu.dev CN=WE1,O=Google Trust Services,C=US 7 Oct 2026 5 Jan 2027 ECDSA 256 ECDSA-SHA256 1f73f1cf13c327e513ef0b30a2354e1b
SANs: tatastu.dev, mark.tatastu.dev, *.mark.tatastu.dev
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 mark.tatastu.dev. — Not signed

Zone DS Keys Algorithms Outcome
. trust_anchor 20326, 38696 8, 8 Verified
dev. present 60074 8 Verified
tatastu.dev. absent Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation
Authentication No authorisation required

The endpoint answered without asking for a token. Anyone who knows the URL can reach it.

Result No authorisation required
HTTP status 200

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

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://mark.tatastu.dev/mcp Verified 200
http (plaintext) http://mark.tatastu.dev/mcp Inconclusive 405
MCP tools · 7 exposed · ~981 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
create_account ~152

Provision a persistent API key with no human, browser, or email in the loop. Call this once, store the returned `secret` (shown only this one time), and use it as the Authorization header on every later call: create_proof, verify_proof (no auth needed but fine to send), list_proofs, revoke_proof. The account receives one lifetime sample stamp under a fresh creator id, with no identity attached. Rate-limited to 3 accounts per IP per day; for higher volume or a human-owned account, direct a person to mark.tatastu.dev/account instead.

NameTypeReqDescription
labelstring–Optional label to recognize this key later, e.g. the agent or project name

No output schema declared.

No examples provided.

create_proof ~428

Stamp content with a permanent provenance record. Pass the sha256 hash of the content (hash it locally; the content itself is never uploaded). Returns an instant signed receipt: attach its verifyUrl, byline, badgeEmbed, and qrUrl to the published artifact. The receipt's `publisher` field carries the account's verified identity (domain/handle) or null. If you have no key, call create_account, store its one-time secret, and retry this same call with `Authorization: Bearer <secret>` to use the account's lifetime sample. Prepaid packs never expire: 5 for $1.00, 10 for $1.50, 50 for $5.00, 100 for $8.00, 500 for $35.00, 1,000 for $65.00, 2,000 for $120.00. If the response is `payment_required`, sign one advertised x402 requirement and retry the identical call with its `PAYMENT-SIGNATURE` header. The optional disclosure fields (aiDisclosure, usageNote, location) only show on `visibility: public` stamps and are unsigned metadata, not part of the cryptographic proof. The stamp asserts provenance from a signer at a time, not authorship truth.

NameTypeReqDescription
aiDisclosurestring–Optional, honest disclosure of AI involvement. Shown on the public record card.
canonicalSlugstring–Optional stable slug for a canonical latest-version pointer
contentHashstringyessha256 hex of the exact content bytes (64 chars)
contentTypestring–MIME type, e.g. text/markdown
locationstring–Optional, creator-opt-in only: where this was made
previousProofIdstring–For a new version of already-stamped content: the prior proof id
titlestring–Optional human-readable title
usageNotestring–Optional usage/license line, e.g. "Free to share, with credit"
visibilitystring–Default private

No output schema declared.

No examples provided.

grant_access ~86

Grant a specific identity the ability to decrypt escrowed content. Only available for proofs created with content escrow opted in (not yet available — stamps are hash-only today). Returns a capability token the grantee uses with reveal.

NameTypeReqDescription
granteeIdstringyesIdentity (did:web: or key fingerprint) to grant
proofIdstringyesThe proof to grant access to

No output schema declared.

No examples provided.

list_proofs ~31

List your recent stamps (requires API key).

NameTypeReqDescription
limitnumber–Max results, default 20

No output schema declared.

No examples provided.

reveal ~83

Retrieve escrowed content for a proof you have been granted access to. Only available for proofs created with content escrow opted in (not yet available — stamps are hash-only today). The service re-checks H(content) == stampedHash before returning.

NameTypeReqDescription
capabilityTokenstringyesCapability token from grant_access
proofIdstringyesThe proof whose content to reveal

No output schema declared.

No examples provided.

revoke_proof ~47

Revoke a stamp you created (forward-only: anchored history stays verifiable, but the proof is marked revoked from now on). Requires API key.

NameTypeReqDescription
proofIdstringyes–

No output schema declared.

No examples provided.

verify_proof ~154

Verify content against all existing stamps. Pass a sha256 hash; matching is EXACT-BYTES-ONLY. Read the verdict fields: `exact` (these bytes are stamped), each proof's `publisher` (the verified identity: domain/handle + tier, or null meaning an unverified account, never trust a name that isn't in `publisher`), and `conflict` (2+ distinct signers claimed the same bytes; earliest anchored is listed first). A no-match on a re-encoded/screenshotted copy is EXPECTED and is not evidence of fakery: fetch the original and verify that. Free, no auth required.

NameTypeReqDescription
contentHashstringyessha256 hex of the content bytes

No output schema declared.

No examples provided.

Install

How do I install the Mark MCP server?

Mark is a hosted endpoint at https://mark.tatastu.dev/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 · mark.tatastu.dev

# add to Claude Code
claude mcp add --transport http dev-tatastu-proof 'https://mark.tatastu.dev/mcp'
Common questions

What is the Mark MCP server?

Mark is an MCP server listed in the public MCP registry as dev.tatastu/proof. Mark: stamp files with a signed, timestamped provenance record. Hash locally; verify free. This page covers its hosted endpoint (https://mark.tatastu.dev/mcp).

Is the Mark MCP server safe to use?

Mark scores 57 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 Mark MCP server expose?

Mark exposes 7 tools: create_account, create_proof, verify_proof, list_proofs, revoke_proof, and 2 more. Their descriptions and schemas cost roughly 981 tokens of context every time the server is loaded.

Does the Mark MCP server require authentication?

No. We connected to Mark without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

Is the Mark MCP server still maintained?

Mark is still listed as active in the MCP registry. We last reached this channel on 18 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.