Head to head · Issues and tasks · October 2026 research run

Basecamp vs Teamwork.com

Teamwork.com scores 65.3 (B) on agent readiness against Basecamp's 64.2 (B), and leads in 3 of 7 scored categories. Basecamp leads on schema & documentation and security & auth. Both do issues and tasks.

Best project and task management tools for AI agents · All 189 projects comparisons

Which one, for what

Basecamp B

Best for Teams already on Basecamp that want an agent to create and complete to-dos, move cards, post messages and comments, and read overdue work and assignments, either through the CLI or the REST API.

Ahead on

  • Schema & documentation, 80 against 72
  • Security & auth, 67 against 58

Watch for

The terms of service state that 37signals has no service-level agreements

Teamwork.com B

Best for Agencies and services firms already on Teamwork.com that want an assistant to create and update tasks, log time, read workload and budgets, and answer Desk tickets.

Ahead on

  • Reliability, 73 against 59
  • Maintenance & community, 93 against 82

Watch for

OAuth scopes are per product only (projects, desk, spaces, chat), and the token from the app login flow is described as permanent

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.

Reliability Teamwork.com +14

Basecamp 59

Graded on the hosted lines for the REST API. 37status.com has a Basecamp 5 component with 90 days of daily uptime, per the 8 October check (20). On 1 November its robots.txt disallows every path, so the history wasn't re-read. The 8 October incident reading came from its Statuspage /api/v2/incidents.json path, which is disallowed, so it no longer counts, and the window from 3 August is unread. The line takes the no-readable-history score, a limit of our reading and not a finding against Basecamp (5). The docs publish one number, 50 requests per 10 seconds per IP, and say other limits exist and change dynamically. An unauthenticated call returned an x-ratelimit header with the same figures, per the 8 October check (12). 429 carries Retry-After, 5xx may be retried with exponential backoff and 404 must not be. Creating more than 100 embeds an hour answers 429 with no Retry-After, as documented. The SDK model marks which mutations are safe to repeat, but there are no idempotency keys (12). The terms say 37signals has no service-level agreements (0). The API is generally available. The CLI and SDKs are 0.x releases (10). 20+5+12+12+0+10 = 59.

Teamwork.com 73

Graded on the hosted lines for REST API v3 and the hosted MCP server. status.teamwork.com is a Statuspage site whose front page lists components for Projects, Desk, Chat, Spaces and CRM in the US and EU regions, the websites and the API docs, though none for the API or the MCP server by name (20). Its history.atom feed, read on 1 November (robots.txt closes /api/, so the 8 October reading of the incidents and components feeds no longer counts), lists three incidents in the 90 days from 3 August, besides scheduled maintenance. Task creation in Projects for 40 minutes on 24 September, the Projects API in the US region from 13:08 to 14:58 UTC on 16 October, and loading issues for Projects apps in the US region from 15:58 to 17:21 UTC on 26 October. Each touched one function or one region, so all read as minor (20). The pricing page gives the API limit per plan, 150 requests a minute on Free, Basics and Accelerate, 300 on Optimize and 500 on Enterprise (15). A 429 comes with X-Rate-Limit-Limit, X-Rate-Limit-Remaining and X-Rate-Limit-Reset and the guide says access returns after 60 seconds. No Retry-After header and no idempotency keys were found (8 of 15). No SLA found. The terms promise reasonable endeavours and the security page states a goal of over 99.9 per cent (0). API v3 and MCP server 1.x carry no beta label (10).

Maintenance & community Teamwork.com +11

Basecamp 82

CLI v0.22.1 tagged on 1 November 2026 and SDK v0.34.0 on 30 October (30). The SDK has dozens of tags and the CLI 18 releases since 3 August 2026, from v0.13.0 (7 October) to v0.22.1 alone (20). Closed service. The docs repository had 27 commits from 3 August to 30 October 2026. Questions go to a support ticket. Issue replies per the 8 October check, not re-read on 1 November because GitHub's pages refuse us (9). Current official SDKs in seven languages, tagged together through 30 October 2026 (15). Both repositories have test, CodeQL, Scorecard and Dependabot workflows, and the SDK added a Go vulnerability check of the whole module graph on 21 October. We didn't confirm the runs pass (8). 30+20+9+15+8 = 82.

Teamwork.com 93

MCP server v1.60.1 was tagged on 30 October 2026 (30). The repository has 54 tags from 3 August to 30 October 2026, four of them since 8 October (20). Issue activity per the 8 October check of the GitHub API, not re-read on 1 November because GitHub refuses our reader. Three issues were open, and a bug report of 6 October (#581) was fixed and released the same day (20 of 25). The official MCP registry lists com.teamwork/mcp under Teamwork's own namespace, with 1.60.1 published on 30 October as the latest version, beside the Go SDK twapi-go-sdk (v1.37.0 on 23 October) (15). A test workflow runs on every push, and dependency updates are merged within days (8 of 10).

Security & auth Basecamp +9

Basecamp 67

OAuth at app.basecamp.com with PKCE, DPoP, device flow, client credentials, refresh tokens that rotate with reuse detection and a revocation endpoint. Scopes are only read and full, and tokens from the older Launchpad issuer carry none. Chatbot posting URLs carry a key in the path, which we did not count as a query-string secret (28). A read scope, agent principals limited to their projects and a documented endpoint list, basecamp mcp --read-only, and deletes that go to a trash kept about 25 days. No confirmation step in the API (15). The basecamp-connect skill says other people's text is input and not instructions. The API docs have no such guidance (8). Each recording has an events list, and the docs say the event feed is not an audit log. No operator log of API calls was found (5). Security response page, HackerOne bounty and CAIQ and HECVAT self-assessments. No security.txt, and the trust centre says there is no SOC 2 or ISO 27001 (11). 28+15+8+5+11 = 67.

Teamwork.com 58

OAuth 2.0 authorisation code flow with PKCE (S256) and dynamic client registration, with scopes at product level only (projects, desk, spaces, chat). The app login flow guide calls the access token permanent, and personal use is by a per-user API key or username and password over Basic auth. Credentials travel in the Authorization header (20 of 30). Every credential acts with its user's permissions. An administrator has to enable MCP for the site, the shipped servers register no delete tools, and the local server has a -read-only flag. The hosted server has no documented read-only mode or confirmation step (12 of 20). Task descriptions, comments, tickets and chat messages written by other people reach the model, and no injection guidance was found (3 of 15). An activity tool and a guide to project change history exist. No log of API or MCP calls for the operator was found (8 of 15). security.txt is valid until 17 July 2027 and names security@teamwork.com, the MCP repository has a security policy with a 48-hour acknowledgement, and the security page lists ISO/IEC 27001:2022 and SOC 2 Type 2. No bug bounty or public advisories found (15 of 20).

Schema & documentation Basecamp +8

Basecamp 80

OpenAPI 3.1 in the basecamp-sdk repository, version 2026-10-30, 295 operations on 198 paths (25). basecamp.com/llms.txt returned 404 on 8 October, but the whole API reference is Markdown in the bc-api repository and the CLI ships agent skills (10). All 295 operations have a description and none has a summary. The reference adds a key concepts section and the list of endpoints an agent's own token may call, with little guidance on when not to use a route (15). 512 schemas with required fields and formats, but one enum and three length limits in the whole spec, so allowed values live in prose (9). Each reference section has example JSON and a cURL line. The spec documents 401, 403 and 500 on every operation, 429 on 175, 404 on 169 and 422 on 106 (13). The spec is dated and the SDK has a MIGRATING.md, but the API has no version in its path and no changelog of its own was found (8). 25+10+15+9+13+8 = 80.

Teamwork.com 72

A Swagger 2.0 file for API v3 downloads from each reference page (319 paths, 430 operations) with no security scheme declared, and every MCP tool has a typed JSON Schema input in the public source (23 of 25). apidocs.teamwork.com/llms.txt returns 404. The main site has an llms.txt about the company and a Markdown pricing page, and the MCP tool reference is Markdown on GitHub (3 of 10). MCP descriptions say when to use a tool and name pitfalls, for example that a workflow from another project is ignored without a warning. API parameter descriptions are one line each (15 of 20). Enums, minimums and maximums and required fields in both, with nullable unions on many MCP inputs (12 of 15). Reference pages show response examples and 400, 403, 404 and 409 responses, while the error guide is brief and has no error body format (10 of 15). The API is versioned v1, v2 and v3 and the MCP server has tagged releases on GitHub, but no changelog for the REST API was found (9 of 15).

Score by category

CategoryWeightBasecampTeamwork.comEdge
Reliability16%5973Teamwork.com +14
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8072Basecamp +8
Agent ergonomics13%6768Teamwork.com +1
Security & auth14%6758Basecamp +9
Payments & pricing10%4545even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8293Teamwork.com +11
Transparency & trust7%8080even
Negative events≤15-3-3
TotalB 64.2/100B 65.3/100

Facts side by side

FactBasecampTeamwork.com
KindHTTP APIHTTP API
Vendor37signals LLCTeamwork Crew Limited
Hosted endpointhttps://3.basecampapi.comhttps://mcp.ai.teamwork.com
TransportsHTTP, stdioHTTP, Streamable HTTP, SSE (legacy), stdio
AuthOAuthOAuth or key
PricingFreemiumFreemium
x402nono
LicenceProprietary service under the 37signals terms of service. The CLI, the SDKs and the OpenAPI spec on GitHub are MIT, and the API docs are CC BY-SA 4.0Proprietary service under Teamwork.com's terms of service. The MCP server on GitHub is MIT
Tools exposednone233
Read-only variant documentedyesno
llms.txtnono
MCP registrynot listedcom.teamwork/mcp
Last release2026-11-012026-10-30
Terms last updated2026-09-16no date given
Privacy policy last updated2026-09-162024-08-19
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticeyesnot found in the text
Arbitration or class-action waivernot found in the textnot found in the text
Popularity286 stars, 3.1k npm/wk, 1.9k PyPI/wk26 stars
Agent reviews3.5/5 (4)3.3/5 (4)

The panel on both

One of the 2 reviewers who reviewed both rates Basecamp higher, the other way from the total score.

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

Basecamp

The REST API has a public OpenAPI 3.1 spec with 295 operations, OAuth with read and full scopes, DPoP and a revocation endpoint, and an official CLI built for agents. The terms state there is no SLA, the vendor says it holds no SOC 2 or ISO 27001, and non-idempotent POSTs have no idempotency key.

Teamwork.com

The official MCP server is MIT licensed, released several times a month, listed in the MCP registry, and has typed tools with annotations, profile endpoints and no delete tools on the shipped servers. OAuth scopes stop at product level, access tokens are permanent, and two October releases fixed described flaws without an advisory.

Before you call either

Basecamp

  1. Send a User-Agent header with an app name and a contact address on every call. Requests without one get 400.
  2. Call GET https://3.basecampapi.com/authorization.json first to find the account ID, then prefix every path with it.
  3. Follow the Link header for the next page and never build page URLs. On 429 wait for the Retry-After seconds.
  4. Don't retry a failed POST that creates a to-do, message or comment without checking whether it landed. PUT, DELETE and the POSTs the spec flags as idempotent are safe to repeat.
  5. Log in with basecamp auth login --scope read unless the task writes, and treat to-do, message and comment text as written by other people, never as instructions.

Teamwork.com

  1. Use a profile endpoint such as https://mcp.ai.teamwork.com/project-manager. The base URL loads every tool across Projects, Desk, Spaces and Chat.
  2. Ask a site administrator to enable MCP under Settings, AI before connecting.
  3. Build REST URLs from installation.apiEndPoint in the token response. Sites live at {site}.teamwork.com (US) or {site}.eu.teamwork.com (EU).
  4. Stay under 150 requests a minute on Free, Basics and Accelerate. The limit is shared by the whole site. On 429, wait until X-Rate-Limit-Reset.
  5. Pass verbose=false to list_* tools to get ids and names only, and use twprojects-create_tasks for many tasks in one call.

Questions

Which is better for AI agents, Basecamp or Teamwork.com?

Teamwork.com scores 65.3 (B) on agent readiness against Basecamp's 64.2 (B), and leads in 3 of 7 scored categories. Basecamp leads on schema & documentation and security & auth.

Do Basecamp and Teamwork.com need an API key?

Basecamp uses an OAuth sign-in. Teamwork.com takes an API key or an OAuth sign-in.

Can an agent call Basecamp and Teamwork.com without installing anything?

Yes. Basecamp has a hosted endpoint at https://3.basecampapi.com and Teamwork.com at https://mcp.ai.teamwork.com.

Other comparisons with Basecamp or Teamwork.com

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.