Microsoft Dynamics 365 Sales

# Fix list: Microsoft Dynamics 365 Sales

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/dynamics-365-sales, the October 2026 research run, assessed 8 October 2026. Grade B, 69.7 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Microsoft Dynamics 365 Sales: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 25 out of 100, up to 9.4 more on the total

Why it scored 25: No x402, MPP or L402 (0). Seat prices are public at $65, $105 and $150 a user a month, paid yearly. MCP calls from agents built outside Copilot Studio are billed in Copilot Credits at published credit counts, and we did not read a dollar price per credit (10). The Dynamics 365 trial page states 30 days free and the Power Apps Developer Plan is a free Dataverse environment with Web API access. Neither page says whether a card is needed (15 of 20). A person registers an app in Entra ID and an administrator creates the application user or allows the MCP client (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Reliability, 65 out of 100, up to 7 more on the total

Why it scored 65: Graded on the Dataverse Web API, with the Dataverse MCP server noted. Service health for a tenant is in the Power Platform and Microsoft 365 admin centres, behind an admin sign-in. The unauthenticated page at status.cloud.microsoft exists, and Microsoft's docs describe it as the place for updates when the admin portals are unavailable. It rendered no components for our reader (10 of 20, a departure from the full 20 because the page with history needs a login). No readable incident history (5). Service protection limits are published per user and web server at 6,000 requests and 20 minutes of execution time per five-minute window, and 52 concurrent requests, plus 40,000 requests per licensed user per 24 hours (15). 429 responses carry Retry-After with retry guidance, and upsert on alternate keys with If-Match gives safe retries for writes (15). The SLA for Microsoft Online Services dated 1 October 2026 pays a 25 per cent credit below 99.9 per cent uptime for Sales Enterprise and Professional (10). Web API v9.2 and the MCP endpoint /api/mcp are generally available (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 3. Security & auth, 74 out of 100, up to 4.6 more on the total

Why it scored 74: OAuth 2.0 through Microsoft Entra ID is the only route, with delegated sign-in or an application user holding a client secret or certificate. The OAuth scopes are coarse (`user_impersonation`, `.default`, `mcp.tools`) and the limits come from Dataverse security roles (26 of 30). Security roles set privileges per table and the MCP server follows them and row-level security. Each MCP client app must be allowed per environment by an administrator, and only Copilot Studio is on by default. The delete tools are described as needing explicit user approval, but Microsoft's sample agent instructions on the configuration page tell the agent not to ask before deleting (17 of 20). Lead, contact and email text written by outsiders reaches the model, and the MCP pages we read give no injection guidance (3 of 15). Dataverse auditing logs record changes and user access, readable through the Web API, once an administrator turns it on. It does not cover retrieve operations (13 of 15). SOC 2 Type 2 lists Dynamics 365 and security.txt points to MSRC and the bounty policy, but the file's Expires date of 23 September 2026 has passed (15 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 4. Schema & documentation, 81 out of 100, up to 3.1 more on the total

Why it scored 81: Each environment serves an OData v4 service document and a CSDL $metadata document at /api/data/v9.2/$metadata, and Learn publishes a table reference. No public OpenAPI document was found (18 of 25). learn.microsoft.com/llms.txt returns 404, but Learn returns Markdown for requests with `Accept: text/markdown`, which is how we read every page (10). The docs state purpose and limits, such as no `$skip` and a 5,000-row page ceiling. The published MCP tool descriptions are one line each (15 of 20). Entity types are typed in CSDL, but queries are OData strings and the MCP `read_query` tool takes a SQL SELECT string (12 of 15). Request and response examples on each page, a status code table with named errors, and a JSON error format with optional detail annotations (13 of 15). The version is in the URL (v9.2), with weekly Dataverse service update release notes. The MCP tool changes are noted without a date (13 of 15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 5. Agent ergonomics, 86 out of 100, up to 2.3 more on the total

Why it scored 86: `$select`, `$top` and `Prefer: odata.maxpagesize` size responses. The MCP server has 15 tools, some of them for schema changes and files (22 of 25). `$filter`, `@odata.nextLink` paging with pages up to 5,000 rows, and no `$skip` (20). Errors are JSON with a code and message, a documented status code table and optional annotations with a help link (17 of 20). PATCH on an alternate key upserts, `If-Match` and `If-None-Match` limit it to update or create, and ETags give optimistic concurrency. The MCP delete tools are described as running only after explicit user approval. We did not read the live tool list, so readOnlyHint and destructiveHint are unconfirmed (15 of 20). Official SDKs for .NET (Microsoft.PowerPlatform.Dataverse.Client 1.2.27) and Python (PowerPlatform-Dataverse-Client 1.1.0). The docs say Microsoft ships no other language libraries for the Web API (12 of 15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 6. Transparency & trust, 75 out of 100, up to 2.2 more on the total

Made of editorial 60, provenance 90.

Why it scored 75: Closed service under the Microsoft Product Terms for Dynamics 365, with MIT SDKs (15). The privacy statement, last updated September 2026, says customer agreements control for enterprise products and points to the Product Terms and the Data Protection Addendum of 22 May 2026, which we did not read. Its retention wording is general, and its line on using data to train AI models is not reconciled with the enterprise terms in what we read (18 of 30). The Power Platform deprecations page was updated on 6 October 2026 with dated notices, and the Web API docs state how versions stay compatible. The MCP tool removals carry no date (15 of 20). The region is chosen per environment, with an EU Data Boundary. We did not read a subprocessor list (12 of 20).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Terms of service: read, states 3 of the 7 things a reader expects (6.6 of 10)
- Privacy policy: read, states 8 of the 8 things a reader expects, and has 1 clause that costs points (8 of 10)
- security.txt: published but past its Expires date (5 of 10)

## 7. Maintenance & community, 79 out of 100, up to 1.8 more on the total

Why it scored 79: Dataverse service update 9.2.26094 went to early release on 2 October 2026, and the Python SDK 1.1.0 was published on 7 October 2026 (30). Service updates are weekly, with nine listed since 7 August, each with release notes (20). Public release notes and community forums. We did not check how issues on the SDK repositories are answered (10 of 15). Current official SDKs for .NET and Python and the npm MCP proxy @microsoft/dataverse 1.0.81 of 26 September 2026. The official MCP registry lists only third-party Dataverse and Dynamics servers (13 of 15). The Python SDK is marked Production/Stable for Python 3.10 and later. CI was not checked because GitHub's API refused us for its rate limit (6 of 10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- unchecked: incident history, because status.cloud.microsoft rendered only a title and tenant service health needs an admin sign-in
- unchecked: whether the 30-day trial or the Power Apps Developer Plan needs a card. The trial sign-up pages we tried on Learn returned 404
- unchecked: the dollar price of a Copilot Credit, which sets the cost of MCP calls from agents built outside Copilot Studio
- unchecked: the text of the Data Protection Addendum of 22 May 2026 and the subprocessor list
- unchecked: CI and issue handling on microsoft/PowerPlatform-DataverseClient-Python, because GitHub's API refused us for its rate limit
- When the MCP tools describe_table, list_tables and fetch were removed, and how much notice was given. The docs note the change without a date
- Whether the live MCP tools set readOnlyHint or destructiveHint, and which transport the remote endpoint speaks
- Whether the SLA covers Sales Premium, which the Sales section heading does not name
- The lead said no Dynamics MCP docs were found. The Dataverse MCP server is documented under Power Apps and is generally available. A Sales-specific MCP page we tried returned 404

## Weaknesses

- No public OpenAPI document. The contract is an OData CSDL $metadata document served by each environment behind sign-in
- A person registers an app in Entra ID and a Power Platform administrator enables it. No keyless route was found
- Service health for a tenant sits behind an admin sign-in, and status.cloud.microsoft showed no components or history to our reader
- MCP calls from agents built outside Copilot Studio are billed in Copilot Credits unless the user holds a Premium or Microsoft 365 Copilot licence
- The MCP tools describe_table, list_tables and fetch were removed and search was renamed. The notice in the docs carries no date

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Send `OData-MaxVersion: 4.0`, `OData-Version: 4.0` and `Accept: application/json`, and name columns with `$select` on every read
- Page with `Prefer: odata.maxpagesize` (up to 5,000) and follow `@odata.nextLink`. `$skip` is not supported
- On 429, wait the seconds in Retry-After. The `x-ms-ratelimit-*` headers are for debugging, not pacing
- Use PATCH on an alternate key to upsert, and add `If-Match: *` when an update must not create a record
- For MCP, call `describe` for logical table and column names before `read_query`. Update any allow list that still names `describe_table`, `list_tables` or `fetch`

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.