TPDOC AI, arm's length automated

Product / File and run the firm · MCP server

Ask about a study from Claude, and get the real answer.

Connect Claude Desktop, or another MCP client, and ask about an engagement or a benchmarking study in plain words. Each answer comes from the same services the app uses, under your own sign-in and role.

MCP client · tpdocai

Sample data

Is the tested party inside the arm’s length range on the Haus Strom Global FZCO study?

get_study_alp{ studyId: "8f3c…e21a" }
range Rule
uae_25_75
iqr Lower
0.1284
iqr Upper
0.255
tested Party Pli
0.1344
risk Classification
WITHIN

The range is 12.84% to 25.50%. At 13.44%, the tested party sits inside it.

Illustrative study, abridged response.

01 / In one line

A question in, a figure from your study out.

The client picks the tool. TAIGA answers from the study, and the client puts it in words.

In one line

Question, tool, figure

Sample data
  1. “Is the tested party inside the range on the Haus Strom distribution study?”
  2. get_study_alp
    range Rule
    uae_25_75
    iqr Lower
    0.1284
    iqr Upper
    0.255
    tested Party Pli
    0.1344
    risk Classification
    WITHIN

74 tools · 10 resources · 8 prompts. Illustrative result.

02 / What it can do

74 tools, grouped by the work.

Read and act on engagements, entities, documents, the IRD wizard and benchmarking studies. 10 resources let a client read an engagement or a study directly, and 8 guided prompts run common jobs like reviewing an extraction or comparing studies.

  • Compliance checks run here too, through the check_compliance tool.

Active tools by area

74 in all

Sample data
  • Standalone benchmarking studiesget_study_alp
    17
  • Document extractiontrigger_extraction
    8
  • Engagement benchmarkingget_benchmarking_status
    8
  • IRD wizardget_wizard_progress
    7
  • Documentsupload_document
    7
  • Entitieslist_entities
    4
  • Engagementslist_engagements
    4
  • Reports and Form 3CEBgenerate_local_file
    6
  • Submission and approvalcheck_submit_readiness
    4
  • Guidance and dashboardsuggest_next_steps
    4
  • Teamslist_my_team_members
    3
  • TP calendarlist_upcoming_deadlines
    1
  • Compliancecheck_compliance
    1

03 / No second copy

Every call goes through the app’s own services.

The MCP server holds no data and keeps no session. Each request is checked, then turned into a call to the same endpoints the web app calls, carrying your firm, user, role and perspective.

No second copy

One request, end to end

Sample data
  1. Your MCP client

    POST /mcp with your sign-in token

  2. TAIGA MCP server

    Verifies the token, checks the tool’s roles, keeps no session and no data

  3. The app’s own services

    GET /api/v1/benchmarking/studies/{studyId}/alp

  4. Your firm’s data

    Scoped to your firm, the same as in the app

Carried on every call

  • Firm
  • User
  • Role
  • Perspective

04 / Roles

Each tool names the roles that may call it.

A tool with no role list is refused by default. Firm admins can call every tool. A refusal comes back as a plain message saying which roles the tool needs.

Roles

Who may call each tool

Sample data
Roles that may call each tool
ToolTaxpayerPractitionerManager
list_engagementsAllowedAllowedAllowed
submit_irdAllowedNot allowedNot allowed
override_comparableNot allowedAllowedAllowed
tpo_approveNot allowedNot allowedAllowed

A taxpayer asks to override a comparable

Error

Role client_user cannot access tool 'override_comparable'. Required: tpo_user | tpo_reviewer

05 / Sign-in

Claude Desktop finds the sign-in on its own.

The server publishes OAuth 2.1 protected-resource metadata, so a client learns where to sign in without setup. Every call then carries a token that is verified before any tool runs.

Sign-in

GET /.well-known/oauth-protected-resource

Sample data
{
  "resource": "https://<your TAIGA API>/mcp",
  "authorization_servers": ["https://<your sign-in host>"],
  "scopes_supported": ["mcp:tools", "mcp:resources", "mcp:prompts"],
  "bearer_methods_supported": ["header"]
}
  1. 1The client reads where to sign in.
  2. 2You sign in there, and the client receives a token.
  3. 3Each call carries your token, checked before any tool runs.
  4. 4No token, or a bad one, gets “Authorization required”.

Host names are placeholders.

Coverage

What it covers.

Tools

74 active

Grouped across 13 areas

Resources

10

Engagements, entities, extractions, benchmarking

Prompts

8

Guided multi-step jobs

Transport

Streamable HTTP

No session kept between requests

Sign-in

OAuth 2.1 discovery

Protected-resource metadata, RFC 9728

Access

Denied by default

Each tool lists the roles that may call it

Questions

Before you ask.

Book a demo

Bring us last year's study.

We'll run it through intake, benchmarking and the report while you watch. Whatever comes out is yours to keep.

Prefer email? Use Contact us.

TPDOC AI uses licensed third-party data internally only and does not resell or provide access to any database. Read the disclaimer.

We use Google Analytics to see which pages are useful. It only loads if you accept, and you can change your choice later from Cookie settings. Privacy policy