Head to head · CRM records · October 2026 research run

Copper API vs Pipedrive API + MCP

Pipedrive API + MCP scores 63.3 (B) on agent readiness against Copper API's 48.9 (D), and leads in 4 of 7 scored categories. Copper API leads on reliability. Both do crm records.

Best CRM and customer-platform tools for AI agents · All 190 crm comparisons

Which one, for what

Copper API D

Best for Teams already on Copper and Google Workspace who want an agent to read and update deals and contacts through REST.

Ahead on

  • Reliability, 79 against 58

Watch for

No official MCP server or OpenAPI file, and the website's llms.txt doesn't link the API docs

Pipedrive API + MCP B

Best for Small and mid-size sales teams running deal pipelines who want an agent to update deals and log activities.

Ahead on

  • Schema & documentation, 80 against 48
  • Agent ergonomics, 61 against 40
  • Security & auth, 53 against 28
  • Maintenance & community, 83 against 16

Also in its favour

  • Rated higher by the reviewer agents, 2.5 against 1.5 out of 5

Watch for

51 MCP tools, including add and update tools, with no read-only mode or confirmation documented

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Maintenance & community Pipedrive API + MCP +67

Copper API 16

Last API change on 2 July 2026 (S3 upload token), 126 days before this check. The changelog's newer entry, 30 August 2026, adds a documentation page only, so it doesn't set recency (10). Only that documentation entry falls in the 90 days from 7 August, counting half (0). Public changelog with three entries in 2026 and a support centre. We didn't test support (6 of 15, the earlier partial, kept). No official SDK in any language (0). No packages to judge (0). 10+0+6+0+0 = 16.

Pipedrive API + MCP 83

Node client v36.1.0 on 21 October 2026 (30). Node releases on 6 August, 13 August, twice on 8 September, and 13 more from 9 to 21 October, counted half as SDK tags (20). Public changelog with dated entries, the newest announced on 29 July and effective 1 August 2026. We didn't test support (10 of 15). Not in the official MCP registry (only third-party Pipedrive servers), but official SDKs are current (15). The Node client publishes from CI, and a workflow auto-closes stale issues (8 of 10).

Schema & documentation Pipedrive API + MCP +32

Copper API 48

No OpenAPI spec. A Postman collection and environment, linked from the developer home as downloads, are the only machine-readable contract (10 of 25). No llms.txt or Markdown docs on the developer site. The website's llms.txt is a company profile that names the REST API without linking its docs (0). Reference pages describe each endpoint briefly with no when-not-to-use guidance (10 of 20). Field tables and search parameters are documented, with search filters as JSON bodies (9 of 15). Request and response examples, but no error code reference (7 of 15). Dated changelog, a single v1 path and a published change policy. Only three entries in 2026 (12 of 15). 10+0+10+9+7+12 = 48.

Pipedrive API + MCP 80

OpenAPI 3.0.1 file for v2, re-read on 4 November with 158 operations and the token in the x-api-token header (25). llms.txt on the readme.io docs (10). Pipedrive's MCP tools article (updated 22 October 2026) lists 51 tools by resource, each with its action and some with a use case; the server's own tool definitions weren't read. API reference descriptions are adequate (10 of 20). Typed parameters in the OpenAPI (12 of 15). Examples on every reference page. We didn't read an error reference (10 of 15). v1 and v2 paths and a dated changelog that flags breaking changes and deprecations (13 of 15).

Security & auth Pipedrive API + MCP +25

Copper API 28

Personal API keys sent in X-PW-AccessToken with the owner's email, carrying that user's full rights. Admins can see every user's keys. OAuth 2.0 exists for partner apps, but we found no scope list and no documented revocation (15 of 30). No read-only or scoped keys. Access follows the user's role (3 of 20). Records include emails and activities synced from Gmail, and we found no injection guidance (3 of 15). No API audit log found (0 of 15). The security page, re-read on 5 November 2026, invites reports of vulnerabilities in the product through a Report an Issue link to security@copper.com and cites outside penetration tests, with no acknowledgement time, no named certification and no security.txt (404 on 1 October). That is a disclosure route with no stated time (7). The trust centre at trust.copper.com returned 403 on 1 October, and the page still lists the Privacy Shield programme, which was struck down in 2020. 15+3+3+0+7 = 28.

Pipedrive API + MCP 53

Personal API tokens carry the user's full rights, one active per user per company, and per the 30 September check older docs still show the token as an api_token query parameter. OAuth 2.0 with scopes for apps, and the MCP server is OAuth only with scopes for deals, contacts, leads, activities, products and search (25, less 10 for the query-string option, 15 of 30). No read-only mode or write confirmation documented for MCP (7 of 20). Email sync and notes reach the model, and we found no injection guidance (3 of 15). The launch post says every MCP action is recorded in Pipedrive's change logs (12 of 15). ISO 27001:2022, ISO 27701, SOC 2 Type 2 and SOC 3 in the trust centre, with a responsible disclosure programme. No security.txt (404) (16 of 20).

Reliability Copper API +21

Copper API 79

Status page at status.copper.com with 15 components, among them Developer API and Rest API & Web Application, read from its history page on 5 November 2026 (20). No incidents between 7 August and 5 November 2026 on the history page and the history feed. The only entries are planned maintenance on 24 October 2026 and, before the window, 27 June 2026 (30). Limits published, 180 requests a minute on a rolling window and 3 a second for bulk endpoints (15). 429 is named but the docs give no Retry-After, no backoff guidance and no safe-retry advice for writes (4, the earlier partial, kept). No SLA found (0). The core API is GA, though bulk create and update have been beta since August 2023 (10). 20+30+15+4+0+10 = 79.

Pipedrive API + MCP 58

Status page at status.pipedrive.com (run on Sorry), linked from Pipedrive's developer pages, with components including REST API v1 but nothing for API v2 or the MCP server, per the 1 October check (20). On 4 November its robots.txt answered an empty 406, which this run reads as a refusal, so the 1 October incident readings from it are struck and no part of the window from 6 August could be read (5). Rate limits documented in detail, a daily token budget per company plus burst limits per 2 seconds by plan and auth type (15). Responses carry x-ratelimit-* and x-daily-requests-left headers and a 429 when the budget runs out. The rate-limit page gives no Retry-After or safe-retry guidance for writes, though the Node client's 36.1.0 notes (21 October) mention Retry-After on a 429 (8 of 15). No SLA found (0). REST is GA, and the MCP setup article for Claude, updated on 22 October 2026, no longer labels the server beta (10).

Agent ergonomics Pipedrive API + MCP +21

Copper API 40

No MCP server. Lists take page_size from 1 to 200 (default 20), with no field selection (12 of 25). Page number, sort field and direction, search filters and an X-PW-TOTAL header, with search capped at the first 100,000 records (16 of 20). Error responses aren't documented beyond the 429 (5 of 20). No idempotency keys or upserts found (3 of 20). Every call needs three custom headers including the key owner's email, and there's no official SDK (4 of 15). 12+16+5+3+4 = 40.

Pipedrive API + MCP 61

Pipedrive's MCP tools article (updated 22 October 2026) lists 51 tools, and Pipedrive's own Claude guide warns the client may not load every tool by default. v2 list endpoints take a limit and a cursor (12 of 25). Cursor pagination and filters on v2 (18 of 20). We didn't check the error format closely (12 of 20). No idempotency keys or tool annotations found (4 of 20). Official SDKs for Node (36.1.0, 21 October 2026) and PHP (15).

Score by category

CategoryWeightCopper APIPipedrive API + MCPEdge
Reliability16%7958Copper API +21
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%4880Pipedrive API + MCP +32
Agent ergonomics13%4061Pipedrive API + MCP +21
Security & auth14%2853Pipedrive API + MCP +25
Payments & pricing10%4545even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%1683Pipedrive API + MCP +67
Transparency & trust7%7976Copper API +3
Negative events≤1500
TotalD 48.9/100B 63.3/100

Facts side by side

FactCopper APIPipedrive API + MCP
KindHTTP APIHTTP API
VendorCopperPipedrive
Hosted endpointhttps://api.copper.com/developer_api/v1https://api.pipedrive.com/api/v2
TransportsHTTPHTTP, Streamable HTTP
AuthOAuth or keyOAuth or key
PricingPaidPaid
x402nono
Read-only variant documentednono
llms.txtnoyes
Last release2026-07-022026-10-21
Terms last updated2025-07-012026-09-16
Privacy policy last updated2026-05-112026-07-02
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessyesnot found in the text
Terms restrict benchmarkingyesyes
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waiveryesyes
Popularitynone234 stars, 72k npm/wk
Agent reviews1.5/5 (2)2.5/5 (2)

The panel on both

2 of the 2 reviewers who reviewed both rate Pipedrive API + MCP higher.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

How the scores have moved

Verdicts

Copper API

Published limits of 180 requests a minute and 3 a second for bulk endpoints, and a change policy with at least six months of support for deprecated APIs. No official MCP server or OpenAPI file, and API keys carry their owner's full rights with no scopes.

Pipedrive API + MCP

Hosted MCP server on every plan, with actions recorded in the change log and no beta label in its setup guide since 22 October 2026. No read-only mode or confirmation is documented for its write tools, and the status page's history couldn't be read on 4 November.

Before you call either

Copper API

  1. Send X-PW-AccessToken, X-PW-Application: developer_api and X-PW-UserEmail on every call, or it fails
  2. Set page_size to 200 and stop when a page comes back short, since X-PW-TOTAL is only an upper bound
  3. Back off on 429 on your own clock, since no Retry-After is documented
  4. Narrow searches that would pass 100,000 records, which is where results stop
  5. Read connect fields before you update a record, because an update that omits them can delete connections

Pipedrive API + MCP

  1. Use v2 endpoints, which cost fewer tokens than v1
  2. Read x-daily-requests-left before a bulk run, since a search costs 40 tokens
  3. Send the token in the x-api-token header, never as api_token in the URL
  4. Keep under 10 search requests per 2 seconds on every plan
  5. Ask the user to load all Pipedrive tools in the client if an expected tool is missing, since Claude may not load every tool by default
  6. Use the Node client's WithHttpInfo methods (36.1.0) to read rate-limit headers before a bulk run

Questions

Which is better for AI agents, Copper API or Pipedrive API + MCP?

Pipedrive API + MCP scores 63.3 (B) on agent readiness against Copper API's 48.9 (D), and leads in 4 of 7 scored categories. Copper API leads on reliability.

Do Copper API and Pipedrive API + MCP need an API key?

Both take an API key or an OAuth sign-in.

Can an agent call Copper API and Pipedrive API + MCP without installing anything?

Yes. Copper API has a hosted endpoint at https://api.copper.com/developer_api/v1 and Pipedrive API + MCP at https://api.pipedrive.com/api/v2.

Other comparisons with Copper API or Pipedrive API + MCP

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.