Payani AI vs Relaticle

Payani AI 13

The server answers 401 without a token, so tools/list and its JSON Schemas were not read. The OAuth and protected-resource metadata are machine-readable, but they describe sign-in, not tools (2 of 25). payani.ai/llms.txt exists but indexes marketing pages, with nothing on the MCP server (3 of 10). The MCP page states what the server is for and what it can't do (no edits, deletes or sends), in eight capability areas rather than per tool (5 of 20). Typed inputs not read (0). The 401 returns a JSON error and error_description and a WWW-Authenticate header naming the resource metadata. No examples of calls (3 of 15). No versioning or changelog found (0).

Relaticle 86

OpenAPI 3.1 at /openapi.json and /openapi.yaml with 34 operations (25). llms.txt with a when-to-use and when-not section, and developer pages (10). The llms.txt and MCP docs say what the API is for and what it doesn't do (no email over REST, no meetings), while only 5 of 34 operations carry a description (14 of 20). 114 required lists and typed MCP inputs, but few enums (11 of 15). 661 examples and responses for 401, 402, 403, 404, 409, 422 and 429 (14 of 15). /v1 and frequent tagged releases, with no changelog page (relaticle.com/changelog is 404) (12 of 15). 25 + 10 + 14 + 11 + 14 + 12 = 86.

Payani AI 35

Read as a closed service. The launch release of 20 October 2026 and the terms and privacy policy of 8 October 2026 are within 30 days (30). No changelog, release notes or dated product updates found, so no count of changes in 90 days (0). Support by email and a form, said to reply usually within one business day, untested, and no public changelog (5 of 25). Not in the official MCP registry (a search for payani returned no servers) and no SDKs (0). No packages to judge (0).

Relaticle 90

v3.8.2 tagged on 5 November 2026 and commits on 6 November (30). Thirteen tags between 9 October and 5 November alone (20). More than 2,500 commits since 1 August and merged pull requests numbered past 1,100; the GitHub API did not answer, so open issues and reply times were not counted (15 of 25). Verified com.relaticle/crm namespace in the official MCP registry with the hosted remote (15). CI, tests, security-audit and Dependabot workflows, PHP 8.5 and Laravel 13 (10). 30 + 20 + 15 + 15 + 10 = 90.

Payani AI 26

Tool count and definition sizes are unknown. The MCP page lists eight read areas, which suggests a small toolset, unconfirmed (10 of 25). No pagination, filters or size controls could be confirmed (0). The 401 message says what to do (connect a Payani account), and nothing else is documented (3 of 20). Every call reads and nothing writes, so retries are safe by design. Annotations unread (10 of 20). Connecting is one URL and an OAuth consent, with no SDK (3 of 15).

Relaticle 59

50 MCP tools, six schema resources and a prompt, with no toolsets or read-only subset (5). Filters combine native fields, custom fields and linked records with $and, $or and $not, plus sort, include, per_page (MCP maximum 25) and cursors (20). Errors come back as HTTP codes and inline MCP error responses such as Invalid ability provided.; no error catalogue was read (12 of 20). Base tool classes set readOnlyHint, destructiveHint, idempotentHint and openWorldHint, and upserts by unique field make creates safe to retry (17 of 20). One URL with OAuth or a token; no official SDK (5 of 15). 5 + 20 + 12 + 17 + 5 = 59.

Payani AI 15

Graded as a hosted MCP server. No status page found, and status.payani.ai did not answer (0). No incident history to read (5). No rate limits found on the MCP page, the help centre or the terms (0). No 429, Retry-After or backoff guidance found (0). The terms of 8 October 2026 say there is no uptime guarantee on any plan (0). Nothing calls the MCP server or Ask Payani beta (10).

Relaticle 40

Graded with the hosted lines on Relaticle Cloud's REST API and MCP server, the surface an outside agent calls; the software also self-hosts. No status page is linked and status.relaticle.com did not connect (0). With no readable history the incident line scores 5. Limits have numbers: 600 a minute per workspace and 300 reads or 60 writes a minute per token in the source, and 120 MCP calls a minute per user in the docs (15). 429 is in the OpenAPI responses and upserts by unique field give a safe retry for creates, but the backoff guidance sits on a docs section we read only as a heading (10 of 15). No SLA (0). Version 3.8, not labelled beta (10). 0 + 5 + 15 + 10 + 0 + 10 = 40.

Payani AI 59

Closed service with terms that name Payani Media LLC in San Francisco and California courts (15). Terms, privacy policy and subprocessors page agree on US storage, 30 days' retention after cancellation then deletion, brief survival in backups and no training of Payani's own models on customer content. They disagree on providers, since the privacy policy says providers aren't permitted to train general-purpose models while the subprocessors page says fal.ai may use anonymised or aggregated inputs. Full account export isn't available yet (18 of 30). Terms changes get 30 days' notice and price changes 30 days, but functions can be added, changed or removed with no stated notice (5 of 20). Subprocessors named with their data and US locations, dated 17 August 2026, with a notification list on request (18 of 20).

Relaticle 73

AGPL-3.0 source with an integration exception, and terms for the Cloud service (30). The privacy policy (4 November 2026) gives retention for server logs (90 days), contact forms (12 months) and paused trial workspaces (deleted after 60 days with notice), and names provider categories; no DPA was found (20 of 30). No deprecation policy or dated notices found (5 of 20). The security page lists each Cloud provider and what it receives, with hosting at Hetzner in Germany, and website analytics through Fathom without CRM content (17 of 20). 30 + 20 + 5 + 17 = 72.

Payani AI 47

OAuth authorisation code with PKCE (S256), dynamic client registration, refresh tokens, a revocation endpoint and a single read scope, with access limited to the workspace the person picks (26 of 30). The whole server is read-only per the MCP page, so an agent has no destructive actions to approve (18 of 20). The data it returns includes contact records, form submissions and website-visitor details that outsiders write, and no injection guidance was found (3 of 15). No audit log or per-call view found in the reviewed pages (0). No security.txt (the path returns the HTML site), no disclosure policy, bug bounty or certification found. The privacy policy states encryption in transit only (0).

Relaticle 57

Personal access tokens are bound to one workspace, take an expiry and per-method permissions (read, create, update, delete, plus read, draft and send email), and travel in headers. OAuth 2.1 through Passport is also offered, but the source says its single mcp:use scope authorises the whole toolset (25 of 30). A read-only token is possible; the docs say MCP writes apply immediately and leave confirmation to the client (10 of 20). Synced email reaches the model through the email tools, and no prompt-injection guidance was found (0). Each record has an activity log of field changes and who made them, and admins see it across the workspace (12 of 15). security.txt valid to May 2027, a disclosure address with a 48-hour acknowledgement target and security-audit workflows in CI; the security page says there is no SOC 2 or ISO 27001 (10 of 20). 25 + 10 + 0 + 12 + 10 = 57.