Security & Testing

stipple.sh

A server that forensically inspects financial documents for authenticity and tampering, classifies their type and origin, and manages verification warrants.

4 endpoints19 known toolsFirst detected July 31, 2026Last detected September 14, 2026

ENDPOINT 1

https://www.stipple.sh/mcp

No auth detected

MCP server metadata

Name: stipple
Version: 1.27.2

Capabilities

experimentalpromptsresourcestools

Server instructions

Free to use: 20 checks a week without an account, 100 with an API key or a signed-in account. Stipple document and evidence checks. Every tool works immediately — no key or signup to start, and every tool here is free within the weekly cap above. Send an API key as 'Authorization: Bearer'. Cheapest call: pass a public `url` string; the server fetches it (nothing to download or encode). Otherwise inline `bytes_b64`. Routing: verify_document = is this file genuine (tamper / AI-generation / arithmetic / provenance). extract_fields = pull values, tables and structure out of a document; its options also classify the type (options.classify) and mask PII (options.redact). verify_identity = Australian 100-point ID check over a document set. check_pack = is an application pack complete. screen_adverse_media = sanctions / negative-news on a person or org (PEP not screened). check_document = sha256 cache-check before verify_document. get_warrant / submit_feedback = retrieve or rate a past result. Results include a `permalink` — a shareable public report to hand to your user. Scores are evidence-backed signals, not verdicts; tools abstain rather than guess. For 'did an AI write this prose' use `detect_ai_text`; for 'are this report's citations real' use `verify_references`. Both are already in this canonical suite; /mcp-aitext and /mcp-verify remain optional compatibility surfaces.

Known tools 14

verify_document

Forensically inspect a document (PDF or image) for authenticity: tampering signs, AI-generation indicators, arithmetic reconciliation (financial docs), and provenance.

Inferred read-only

check_document

Cheap cache-check: has this exact document already been inspected?

Inferred read-only

get_warrant

Retrieve a stored warrant by id (e.g.

Inferred read-only

submit_feedback

Record thumbs up/down on a warrant's rating (the engine's precision-flywheel label source).

Inferred read-only

extract_fields

Extract structured FIELDS from a document (PDF or image) with a vision model.

Inferred read-only

verify_identity

Run an Australian identity check over a SET of identity documents.

Inferred read-only

check_pack

Check whether a SET of documents satisfies a checklist — completeness, cheaply.

Inferred read-only

screen_adverse_media

Screen a person or organisation for ADVERSE MEDIA and SANCTIONS exposure (KYC/AML).

Inferred read-only

detect_ai_text

Estimate the PROBABILITY that a document's text was AI-GENERATED (LLM-written prose).

Inferred read-only

verify_references

Fact-check a document's REFERENCES and CLAIMS — built for AI-generated reports whose citations must be checked before they're trusted.

Inferred read-only

check_source_overlap

Check whether text OVERLAPS text published on the public web — a plagiarism-style check: does this text appear elsewhere?

Inferred read-only

find_tenders

Search open tenders across Australia and New Zealand.

Inferred read-only

match_tenders

Rank open tenders against what a company actually does.

Inferred read-only

tender_sources

Every source we search, what it is allowed to do, and what the last run returned.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.stipple]
url = "https://www.stipple.sh/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "stipple": {
      "type": "http",
      "url": "https://www.stipple.sh/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: stipple
Remote MCP URL: https://www.stipple.sh/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "stipple": {
      "url": "https://www.stipple.sh/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "stipple": {
      "type": "http",
      "url": "https://www.stipple.sh/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "stipple",
  "transport": "streamable-http",
  "url": "https://www.stipple.sh/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

ENDPOINT 2

https://www.stipple.sh/mcp-aitext

No auth detected

MCP server metadata

Name: stipple-aitext
Version: 1.27.2

Capabilities

experimentalpromptsresourcestools

Server instructions

AI-text detection. Works immediately — no key or signup to start: free credits weekly per IP, each call costs credits (this tool: 1 PER PROSE SECTION — a short paste is one, long text is split and scored in pieces, and an abstention costs nothing; table at GET /v1/pricing), a free API key gives you an allowance of your own instead of one shared per IP, and holds purchased credits. detect_ai_text estimates the probability prose was LLM-written and returns the specific linguistic tells behind the score. Cheapest call: pass pasted `text` or a public `url` string. The probability is a confidence signal, not proof, and the tool abstains (applicable=false) on forms/tables/scans where style detection is unreliable. For document forensics (tampering, fake payslips) use the /mcp server; for citation fact-checking use /mcp-verify.

Known tools 1

detect_ai_text

Estimate the PROBABILITY that a document's text was AI-GENERATED (LLM-written prose).

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.stipple-aitext]
url = "https://www.stipple.sh/mcp-aitext"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "stipple-aitext": {
      "type": "http",
      "url": "https://www.stipple.sh/mcp-aitext"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: stipple-aitext
Remote MCP URL: https://www.stipple.sh/mcp-aitext

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "stipple-aitext": {
      "url": "https://www.stipple.sh/mcp-aitext"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "stipple-aitext": {
      "type": "http",
      "url": "https://www.stipple.sh/mcp-aitext"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "stipple-aitext",
  "transport": "streamable-http",
  "url": "https://www.stipple.sh/mcp-aitext"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

ENDPOINT 3

https://www.stipple.sh/mcp-verify

No auth detected

MCP server metadata

Name: stipple-verify
Version: 1.27.2

Capabilities

experimentalpromptsresourcestools

Server instructions

Citation fact-checking. Works immediately — no key or signup to start: free credits weekly per IP, each call costs credits (this tool: 2, deep 5; table at GET /v1/pricing), a free API key gives you an allowance of your own instead of one shared per IP, and holds purchased credits. verify_references checks a report's citations (do they resolve? are they the claimed source? in deep mode, do they support the claim?), recomputes its internal maths, and flags unsupported claims. Built for AI-generated reports. Cheapest call: pass a public `url` string. Returns a shareable `permalink` report to hand to your user. Reports coverage, not truth. For document forensics use the /mcp server; for 'did an AI write this prose' use /mcp-aitext.

Known tools 1

verify_references

Fact-check a document's REFERENCES and CLAIMS — built for AI-generated reports whose citations must be checked before they're trusted.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.stipple-verify]
url = "https://www.stipple.sh/mcp-verify"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "stipple-verify": {
      "type": "http",
      "url": "https://www.stipple.sh/mcp-verify"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: stipple-verify
Remote MCP URL: https://www.stipple.sh/mcp-verify

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "stipple-verify": {
      "url": "https://www.stipple.sh/mcp-verify"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "stipple-verify": {
      "type": "http",
      "url": "https://www.stipple.sh/mcp-verify"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "stipple-verify",
  "transport": "streamable-http",
  "url": "https://www.stipple.sh/mcp-verify"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

ENDPOINT 4

https://www.stipple.sh/mcp-tenders

No auth detected

MCP server metadata

Name: stipple-tenders
Version: 1.27.2

Capabilities

experimentalpromptsresourcestools

Server instructions

Open Australian and New Zealand government tenders, and capability matching against a company's own website. Works immediately — no key or signup: free credits weekly per IP, and a free API key gives you an allowance of your own. find_tenders and tender_sources are FREE; match_tenders costs 3 credits because it reads a company site and scores a shortlist against it. Every response carries `coverage`: which sources were searched and which returned nothing. An empty result means no match in what we searched, never that no such tender exists.

Known tools 3

find_tenders

Search open government tenders across Australia and New Zealand.

Inferred read-only

match_tenders

Rank open tenders against what a company actually does.

Inferred read-only

tender_sources

Every source we search, what it is allowed to do, and what the last run returned.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.stipple-tenders]
url = "https://www.stipple.sh/mcp-tenders"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "stipple-tenders": {
      "type": "http",
      "url": "https://www.stipple.sh/mcp-tenders"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: stipple-tenders
Remote MCP URL: https://www.stipple.sh/mcp-tenders

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "stipple-tenders": {
      "url": "https://www.stipple.sh/mcp-tenders"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "stipple-tenders": {
      "type": "http",
      "url": "https://www.stipple.sh/mcp-tenders"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "stipple-tenders",
  "transport": "streamable-http",
  "url": "https://www.stipple.sh/mcp-tenders"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Trust Data Available

BuiltWith Trust API v2 evidence for stipple.sh was fetched 2026-08-21T14:30:03.063Z and is being refreshed.

Trust status Neutral

stipple.sh is assessed as Neutral: No suspicious signals found, but no strong positive signal either

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.