Untax
REMOTE · UNTAX.IO
Source-backed property assessment research and draft assessment comparisons for supported pilot j...
Available components
Last scanned (9 hours ago)
Scores higher than 55% of remote servers
Reachable. Last reached 5 Nov 2026.
Category Finance and Payments
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 Security66
- The endpoint's TLS certificate is valid, in date, and uses a strong key. View diagnostics → Pass
- Authorisation check failed: no authorisation is required to call this server, and it exposes a tool marked destructive (disconnect_va_connection). See how to fix → View diagnostics → Fail
- HTTPS is enforced; there's no plaintext access path. View diagnostics → Pass
- The HSTS (Strict-Transport-Security) header is present. View diagnostics → Pass
- DNSSEC is configured correctly; the domain's records validate against the full chain to the root. View diagnostics → Pass
Transport & Reachability100
- Verified streamable-http transport via a live MCP handshake. View diagnostics → Pass
Schema Quality & AI Usability76
- 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 4037 tokens (~192/item across 21 items; 20 tools + 1 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 Management23
- Stability check failed: schema churn in the 10 days we've observed: 2 tool removals, 0 breaking changes, 0 auth/transport breaks, 2 additions. See how to fix → Fail
Tool Coverage82
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 45% 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
- We read all 20 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 22 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a current MCP spec version (2026-07-28).Pass
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.
- 5 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 19 to 23.
- 3 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 13 to 16.
- 1 Nov 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 6 to 9.
- 30 Oct 26 −1
- Stability: 0.10 → fail ▼ security
- A breaking change shipped without a version bump: still 0.4.1 ▼ security
- Tool “audit_microtransaction” was removed ▼ security
- Tool “get_microtransactions_catalog” was removed ▼ security
- New tool “list_property_services” functional
- New tool “review_property_service” functional
- 29 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.
- 27 Oct 26 +2
- DNSSEC: fail → pass ▲ security
- The server rewrote its instructions, which are the text every model session reads security
- Tool “audit_microtransaction” rewrote its description, which is the text the model reads security
- Tool “complete_tax_sandbox_order” rewrote its description, which is the text the model reads security
- Tool “disconnect_va_connection” rewrote its description, which is the text the model reads security
- Tool “find_va_facilities” rewrote its description, which is the text the model reads security
- Tool “find_va_forms” rewrote its description, which is the text the model reads security
- Tool “generate_veteran_packet” rewrote its description, which is the text the model reads security
- Tool “get_appeal_lifecycle” rewrote its description, which is the text the model reads security
- Tool “get_microtransactions_catalog” rewrote its description, which is the text the model reads security
- Tool “get_owner_review_details” rewrote its description, which is the text the model reads security
- Tool “get_owner_workspace” rewrote its description, which is the text the model reads security
- Tool “get_property_assessment” rewrote its description, which is the text the model reads security
- Tool “get_tax_sandbox_order” rewrote its description, which is the text the model reads security
- Tool “get_va_connection” rewrote its description, which is the text the model reads security
- Tool “get_va_facility” rewrote its description, which is the text the model reads security
- Tool “get_va_form” rewrote its description, which is the text the model reads security
- Tool “recheck_property_opportunities” rewrote its description, which is the text the model reads security
- Tool “review_veteran_requirements” rewrote its description, which is the text the model reads security
- Tool “start_tax_sandbox” rewrote its description, which is the text the model reads security
- Tool “start_va_connection” rewrote its description, which is the text the model reads security
- Tool “suggest_addresses” rewrote its description, which is the text the model reads security
- Stability: unverified → 0.03 ▲ functional
- “get_property_assessment” reworded the description of “actual_sqft” cosmetic
- “get_property_assessment” reworded the description of “county_key” cosmetic
- “recheck_property_opportunities” reworded the description of “county_key” cosmetic
- 26 Oct 26 65
First indexed and scored.
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 5 Nov 2026 · Probed https://untax.io/api/v1/mcp
TLS valid
Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .
| Subject | Issuer | Valid from | Valid until | Key | Signature | Serial |
|---|---|---|---|---|---|---|
| CN=untax.io | CN=YE2,O=Let's Encrypt,C=US | 24 Sept 2026 | 23 Dec 2026 | ECDSA 256 | ECDSA-SHA384 | 63c1413a3dd178fc74ce1913fde8716b2ba |
| SANs: *.untax.io, untax.io | ||||||
| CN=YE2,O=Let's Encrypt,C=US (CA) | CN=Root YE,O=ISRG,C=US | 3 Sept 2025 | 2 Sept 2028 | ECDSA 384 | ECDSA-SHA384 | 4df3b15dd6c0784c507cd37b58e6f115 |
| CN=Root YE,O=ISRG,C=US (CA) | CN=ISRG Root X2,O=Internet Security Research Group,C=US | 13 May 2026 | 2 Sept 2032 | ECDSA 384 | ECDSA-SHA384 | 872165fc34b6e5fba8add5b3705fb53a |
| CN=ISRG Root X2,O=Internet Security Research Group,C=US (CA) | CN=ISRG Root X1,O=Internet Security Research Group,C=US | 13 May 2026 | 2 Sept 2032 | ECDSA 384 | SHA256-RSA | 6c8f1dc727c7117f7baf853ac980f9cd |
Background: What to check on a remote MCP endpoint →
DNSSEC secure
Validation of untax.io. — Secure
| Zone | DS | Keys | Algorithms | Outcome |
|---|---|---|---|---|
| . | trust_anchor | 20326, 38696 | 8, 8 | Verified |
| io. | present | 38690 | 8 | Verified |
| untax.io. | present | 2371 | 13 | Verified |
| untax.io. | Verified address RRset verified with the apex keys |
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 |
| Header | Value |
|---|---|
| strict-transport-security | max-age=63072000; includeSubDomains; preload |
| content-security-policy | default-src 'self'; script-src 'self' 'unsafe-inline' https://challenges.cloudflare.com https://unpkg.com; style-src 'self' 'unsafe-inline' https://fonts.googleapis.com; font-src 'self' https://fonts.gstatic.com; img-src 'self' data: https:; connect-src 'self' https://untax.io; frame-src https://challenges.cloudflare.com; frame-ancestors 'none'; |
| x-content-type-options | nosniff |
| x-frame-options | DENY |
| referrer-policy | no-referrer |
| permissions-policy | camera=(), microphone=(), geolocation=() |
Background: How OAuth 2.1 works in the 2026 MCP spec →
Transports 2 probes
| Transport | URL | Outcome | Status | Location |
|---|---|---|---|---|
| streamable-http | https://untax.io/api/v1/mcp | Verified | 200 | |
| http (plaintext) | http://untax.io/api/v1/mcp | HTTPS enforced |
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 →
complete_tax_sandbox_order ~213
Check the test payment for your own order, then make its draft PDF and receipt available if verified. Requires owner:sandbox sign-in and completed test checkout. Any submission is simulated. It does not file a real appeal or change county records. Show the returned test checkout link when payment is pending. When an order_id is known, read the same order after checkout or an uncertain failure before retrying. If starting checkout fails before an order_id is returned, retry start_tax_sandbox with the original idempotency_key; do not create a new key for the same attempt. get_tax_sandbox_order and complete_tax_sandbox_order take only order_id. Report test payment, draft generation, download availability and external delivery as separate observed states. Do not claim the draft is available until the order confirms its artifact is available; a payment or download URL alone is not proof. A sandbox receipt or simulated submission is not county filing, external email delivery or permission for live payments.
| Name | Type | Req | Description |
|---|---|---|---|
| order_id | string | yes | – |
No output schema declared.
No examples provided.
disconnect_va_connection ~57
When requested, stop this assistant's access to this account's shared VA facts. Requires owner:va. Saved verification stays in Account until the owner removes it. Does not disconnect another account or assistant, submit a claim or change VA records.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
find_va_facilities ~96
Find official VA facilities in a specified state, with optional city and facility-type filters. If location is missing, ask the returned question. Do not guess a state or present a VA facility as a county tax office or accredited VSO. Results are public information; sandbox results are samples.
| Name | Type | Req | Description |
|---|---|---|---|
| city | string | – | – |
| facility_type | string | – | – |
| limit | integer | – | – |
| state | string | – | – |
No output schema declared.
No examples provided.
find_va_forms ~64
Find official VA forms by number or search text. If query is missing, ask which form or purpose is needed. Show returned revision details and official PDF links. This does not complete or submit a claim.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | – |
| query | string | – | – |
No output schema declared.
No examples provided.
generate_veteran_packet ~165
Create a free private veteran review PDF for the selected saved review. Requires owner:read. Missing answers appear as questions to review. A VA connection is optional; using connected facts requires owner:va and the user's explicit choice for this applicant. Extra answers are user reports and are not saved to the review. Returns a protected, expiring PDF download. The draft does not complete an official application, decide eligibility, estimate savings or file anything.
| Name | Type | Req | Description |
|---|---|---|---|
| answers | object | – | – |
| review_id | string | yes | – |
| saved_property_id | string | yes | – |
| use_connected_va_records | boolean | – | Only true when the user explicitly asks to use their connected records for their own applicant review; requires additional owner:va permission and is_surviving_spouse=false. |
No output schema declared.
No examples provided.
get_appeal_lifecycle ~50
Read a general checklist for reviewing an assessment. It does not look up a property or verify deadlines, fees, eligibility or outcomes. Check the official rules for the property's location and tax year.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
get_owner_review_details ~101
Read the selected saved property review. Requires both IDs from get_owner_workspace and valid owner:read permission. Check source_context_current and reconfirmation_required; confirm stale previous_owner_answers with the user before reusing them. Ask relevant unanswered questions and honor current answers and skips. Keep user reports separate from county facts. Use the protected browser review for changes and sensitive evidence.
| Name | Type | Req | Description |
|---|---|---|---|
| review_id | string | yes | – |
| saved_property_id | string | yes | – |
No output schema declared.
No examples provided.
get_owner_workspace ~66
Read saved property reviews for the signed-in Untax account. Requires valid, revocable owner:read permission. Returns property and review IDs; do not assume an active property. Ask the user to choose when ambiguous. This does not verify ownership, benefit eligibility, savings or filing authority.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
get_property_assessment ~435
Read county assessment records and their sources for a supported address or parcel ID. Keep missing facts unknown; commercial building area is not finished living area. Report comparisons as drafts with their evidence limits. They do not verify personal tax savings or authorize filing. For a property audit or a request to find tax-reduction routes, review opportunity_screen before concluding. A weak or held comparable-property analysis does not rule out independent relief, record corrections or billing research. Ask the next relevant unanswered targeted_questions item one at a time; use available_questions for relevant follow-ups. Accept none, not sure or skip, and do not repeat answered or skipped questions in this conversation unless the user changes the facts. After an answer, call recheck_property_opportunities for the same selected property and county with the accumulated explicit owner_answers and skipped_question_ids. Rechecks are stateless: resend still-applicable prior answers, preserve explicit false and empty arrays, and omit unknown or skipped fields instead of sending null or inventing a negative answer. Resend stable skipped_question_ids until the user supplies an answer or asks to revisit the question. Skips suppress prompts only and leave requirements unresolved. Do not transfer property-specific answers or skips to another property. When the user is done or skips the remaining questions, summarize supported research routes and missing evidence without treating unknowns as a denial. For a fact-only lookup, answer the requested fact and offer relevant screening without requiring an unrelated intake. Owner answers do not verify eligibility, savings, authority, payment or filing readiness.
| Name | Type | Req | Description |
|---|---|---|---|
| actual_sqft | integer | – | Optional owner-reported measured living area, in square feet |
| address | string | – | Alternative alias for property street address |
| address_or_pidn | string | yes | Full street address or county parcel ID (e.g. '3731 Anderson Ave, Chattanooga, TN' or '157I L 001') |
| county_key | string | – | Optional registered county key, such as hamilton. Searches only that county. A conflicting county prefix is rejected. |
| notice_appraised_value | number | – | Optional: New proposed appraised value from reassessment notice |
No output schema declared.
No examples provided.
get_tax_sandbox_order ~198
Read your own Untax test order. Requires owner:sandbox sign-in. Reports test payment and draft delivery separately. Neither status verifies production eligibility or a county filing. Show the returned test checkout link when payment is pending. When an order_id is known, read the same order after checkout or an uncertain failure before retrying. If starting checkout fails before an order_id is returned, retry start_tax_sandbox with the original idempotency_key; do not create a new key for the same attempt. get_tax_sandbox_order and complete_tax_sandbox_order take only order_id. Report test payment, draft generation, download availability and external delivery as separate observed states. Do not claim the draft is available until the order confirms its artifact is available; a payment or download URL alone is not proof. A sandbox receipt or simulated submission is not county filing, external email delivery or permission for live payments.
| Name | Type | Req | Description |
|---|---|---|---|
| order_id | string | yes | – |
No output schema declared.
No examples provided.
get_va_connection ~103
Read the limited, dated VA facts shared with this account and assistant. Requires owner:va and active sharing permission; browser sign-in alone grants no access. Saved facts remain after the VA session expires or the server restarts. Show verified_at and freshness. This is a saved snapshot, not a live check; reverification may be needed. Expired or revoked app permission stops sharing. These facts do not establish property ownership, primary residence, tax eligibility or filing authority.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
get_va_facility ~51
Read one official VA facility using its returned id. Returns available address, phone and website details, with source and environment. Does not return private veteran records or decide eligibility.
| Name | Type | Req | Description |
|---|---|---|---|
| id | string | – | – |
No output schema declared.
No examples provided.
get_va_form ~107
Look up a VA form number, such as 21-0966. Use form_name from search results, not the numeric catalog id. If form_name is missing, ask for the form number. Returns available title, revision details and official PDF link. The link does not complete or submit an application or establish eligibility.
| Name | Type | Req | Description |
|---|---|---|---|
| id | string | – | Official form number (the form_name field in search results), such as 21-0966. Do not use the numeric catalog id. |
No output schema declared.
No examples provided.
list_property_services ~54
List available property-tax research services, their current status and required evidence. Use the returned service identifier when reviewing a selected property. Results do not determine eligibility or savings, set a payable price, or authorize a refund or filing.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
recheck_property_opportunities ~410
Review possible property-tax options using the available county records and the owner's answers. Returns research and next questions. Answers do not verify eligibility, evidence, savings, owner authority or permission to pay or file. Do not send exact income, ages, identifiers, documents or a caller-supplied audit. For a property audit or a request to find tax-reduction routes, review opportunity_screen before concluding. A weak or held comparable-property analysis does not rule out independent relief, record corrections or billing research. Ask the next relevant unanswered targeted_questions item one at a time; use available_questions for relevant follow-ups. Accept none, not sure or skip, and do not repeat answered or skipped questions in this conversation unless the user changes the facts. After an answer, call recheck_property_opportunities for the same selected property and county with the accumulated explicit owner_answers and skipped_question_ids. Rechecks are stateless: resend still-applicable prior answers, preserve explicit false and empty arrays, and omit unknown or skipped fields instead of sending null or inventing a negative answer. Resend stable skipped_question_ids until the user supplies an answer or asks to revisit the question. Skips suppress prompts only and leave requirements unresolved. Do not transfer property-specific answers or skips to another property. When the user is done or skips the remaining questions, summarize supported research routes and missing evidence without treating unknowns as a denial. For a fact-only lookup, answer the requested fact and offer relevant screening without requiring an unrelated intake. Owner answers do not verify eligibility, savings, authority, payment or filing readiness.
| Name | Type | Req | Description |
|---|---|---|---|
| address_or_pidn | string | yes | – |
| county_key | string | – | Optional registered county key, such as hamilton. Searches only that county. A conflicting county prefix is rejected. |
| owner_answers | object | yes | – |
| skipped_question_ids | array | – | Stable question IDs explicitly skipped in this same property conversation. Resend them with accumulated answers. Skips suppress prompts only; facts and eligibility remain unresolved. |
No output schema declared.
No examples provided.
review_property_service ~196
Review a selected property's source facts and missing evidence for one requested property-tax research service. Choose service_type from list_property_services. The result describes available research actions; it does not verify savings, quote a payable fee, charge, prepare a filing packet or submit a correction. Test checkout is a separate account workflow.
| Name | Type | Req | Description |
|---|---|---|---|
| address | string | – | Explicitly selected property street address or parcel ID. A nonblank address takes precedence over parcel_id. |
| county_key | string | – | Exact registered county adapter key, such as hamilton; restricts source lookup to that county. |
| parcel_id | string | – | Explicitly selected county parcel ID / PIDN, used when address is absent or blank. |
| service_type | string | yes | Explicit research service selected from list_property_services. Accepted aliases include demolished_outbuilding, square_footage_correction, mailing_address_change, commercial_reclassification and cer… |
No output schema declared.
No examples provided.
review_veteran_requirements ~176
Review requirements for the selected property using accumulated answers and, only when requested, the connected applicant's saved VA facts. Requires owner:va. Show va_source.verified_at and freshness; saved facts are not a live check. Ask returned questions one at a time. Accept unknown or skip, preserve explicit false, and keep answers with the same property. Does not decide eligibility or savings, generate an application, or file anything.
| Name | Type | Req | Description |
|---|---|---|---|
| answers | object | – | – |
| skipped_fields | array | – | Stable question ids explicitly skipped in this same property conversation. Unknown facts stay unresolved. |
| use_connected_va_records | boolean | – | Set true only when the user explicitly identifies the connected VA records as theirs and asks to use them for their own applicant review. A surviving-spouse review needs separate deceased-veteran and… |
No output schema declared.
No examples provided.
start_tax_sandbox ~244
Start a test checkout for the selected property only when the user requests it. Requires owner:sandbox sign-in. Uses test payments and separate test orders. It does not verify ownership, eligibility, savings or an open appeal window. No real charge or county filing occurs. Show the returned test checkout link when payment is pending. When an order_id is known, read the same order after checkout or an uncertain failure before retrying. If starting checkout fails before an order_id is returned, retry start_tax_sandbox with the original idempotency_key; do not create a new key for the same attempt. get_tax_sandbox_order and complete_tax_sandbox_order take only order_id. Report test payment, draft generation, download availability and external delivery as separate observed states. Do not claim the draft is available until the order confirms its artifact is available; a payment or download URL alone is not proof. A sandbox receipt or simulated submission is not county filing, external email delivery or permission for live payments.
| Name | Type | Req | Description |
|---|---|---|---|
| address_or_pidn | string | yes | – |
| county_key | string | yes | – |
| idempotency_key | string | yes | – |
| service | string | yes | – |
No output schema declared.
No examples provided.
start_va_connection ~96
Start a requested VA connection for this signed-in account and assistant. Requires owner:va. Show connection_url to confirm the account and fields to share. Review saved verification or complete official VA sign-in, then approve sharing with this app. Verification is saved to the account only after approval. Call get_va_connection after browser return. The link alone grants no access. Never request passwords, tokens, SSNs or full VA documents in chat.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
suggest_addresses ~116
Find possible address matches in county records from a partial query or misspelling. If the property is ambiguous, ask the user to choose before drawing conclusions or taking action. Similarity scores and is_auto_match are hints. They do not verify a selection, owner authority or financial eligibility.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | Maximum number of candidate suggestions to return (default: 5) |
| query | string | yes | Address query or partial street name (e.g. '3731 Andrson Ave', 'Anderson Ave', '810 Lindsay') |
No output schema declared.
No examples provided.
How do I install the Untax MCP server?
Untax is a hosted endpoint at https://untax.io/api/v1/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 · untax.io
claude mcp add --transport http io-untax-untax 'https://untax.io/api/v1/mcp'
What is the Untax MCP server?
Untax is an MCP server listed in the public MCP registry as io.untax/untax. Source-backed property assessment research and draft assessment comparisons for supported pilot j... This page covers its hosted endpoint (https://untax.io/api/v1/mcp).
Is the Untax MCP server safe to use?
Untax scores 70 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 Untax MCP server expose?
Untax exposes 20 tools: suggest_addresses, get_property_assessment, recheck_property_opportunities, get_owner_workspace, get_owner_review_details, and 15 more. Their descriptions and schemas cost roughly 2,998 tokens of context every time the server is loaded.
Does the Untax MCP server require authentication?
No. We connected to Untax without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.
Is the Untax MCP server still maintained?
Untax is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.