Head to head · Lead enrichment · October 2026 research run

Databar vs Gojiberry

Databar scores 53.8 (D) on agent readiness against Gojiberry's 48.2 (D), and leads in 5 of 7 scored categories. Gojiberry leads on reliability and transparency & trust. Both do lead enrichment.

Best lead, contact and company data APIs for AI agents · All 231 leads comparisons

Which one, for what

Databar D

Best for Waterfall enrichment across many providers for GTM agents, close to clay and fullenrich.

Ahead on

  • Schema & documentation, 84 against 71
  • Agent ergonomics, 66 against 55
  • Security & auth, 47 against 31
  • Payments & pricing, 35 against 10
  • Maintenance & community, 73 against 20

Also in its favour

  • Free to start without a card

Watch for

API access needs a paid plan; the free plan's 150 credits don't include it

Gojiberry D

Best for A founder or small sales team that wants an agent to pull scored, signal-based B2B leads and run LinkedIn and email outreach from one paid workspace.

Ahead on

  • Reliability, 70 against 28
  • Transparency & trust, 61 against 55

Watch for

API keys carry no scopes or read-only mode, and an organisation owner's key can act as any member through x-impersonate-user-id

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 Databar +53

Databar 73

Closed service. Python SDK 2.9.0 on 30 October 2026, 7 days before this check (30). SDK 2.7.0 and 2.8.0 on 6 October and changelog entries on 23, 24, 27, 28, 29 and 30 September and 1 October (20). A public changelog, roadmap and feedback portal, an email address, and a Slack channel on Scale (10 of 15). An official Python SDK and CLI, no other languages (10 of 15). Package health not checked (3 of 10). 30+20+10+10+3 = 73.

Gojiberry 20

No changelog or dated API release was found. The developer docs carry no dates, and their sitemap stamps record no change. The newest dated evidence is help-centre editing. The MCP article is dated 29 September 2026 and the API article 21 August 2026, so the partial credit for activity inside 90 days is kept (15 of 30). No dated releases to count (0). Live chat and hello@gojiberry.ai for support, not tested, and no public changelog (5 of 15). No readable registry page names the server and no official SDKs exist (0). Package health 5 to 0, an in-between value with no package behind it. 15 + 0 + 5 + 0 + 0 = 20.

Reliability Gojiberry +42

Databar 28

Hosted lines. No status page linked from the site or docs (0). No incident history to read (5). The 20 August changelog says rate limits return 429, but no numbers were found in the docs read (5 of 15). The SDK backs off on 429 per the 28 September entry, tasks can be cancelled with a refund, and no Retry-After or idempotency key is documented (8 of 15). No SLA found (0). The MCP server left beta on 15 July 2026 and the API is generally available (10). 0+5+5+8+0+10 = 28.

Gojiberry 70

Graded with the hosted lines. Status page at status.gojiberry.ai on Better Stack with two components, gojiberry.ai and api.gojiberry.ai, and 90-day bars. Neither names ext.gojiberry.ai or mcp.gojiberry.ai, the hosts an agent calls, so the page has no component for the product (15, was 20). Both show 100 per cent, and the incident history lists none for September, October or November 2026 (August per the 8 October check). The page has not been shown to post any incident, so 20 rather than a clean record (was 22, an in-between value; tldv, zamp). The quickstart and the OpenAPI file state 100 requests a minute and 100,000 a day per account, shared by all keys and MCP connections, with lower limits on some endpoints (15). 429 line 7 to 10. Requests over a limit now get 429 Too Many Requests with a Retry-After header and the instruction to wait that long before retrying. No idempotency keys or safe-retry guidance for writes, so Retry-After alone (10). No SLA found, and the terms say the service is supplied as is (0). API version 1.0 with no beta label (10). 15 + 20 + 15 + 10 + 0 + 10 = 70.

Payments & pricing Databar +25

Databar 35

No x402, MPP or L402 (0). Plan prices and credits are public, and each connector's credit cost is shown before a run and in the enrichment's pricing field; credits aren't sold per call outside a plan (15 of 20). A 14-day premium trial with no card, per the pricing FAQ (20). A person signs up in a browser, and API access needs a paid plan (0). 0+15+20+0 = 35.

Gojiberry 10

No x402, MPP or L402 (0). The Pro plan is public at $99 a month, and credit costs per action are published, but no dollar price per credit was found (10). There's a free trial, and the help centre says the subscription then starts on the saved payment method, so we read it as needing a card. We didn't open the signup form (0). A person signs up in a browser and the terms forbid account creation by bots (0). 0 + 10 + 0 + 0 = 10.

Security & auth Databar +16

Databar 47

A workspace API key in a header, never in the URL, and OAuth on the MCP with a Connected apps page to revoke access. No scopes found (20 of 30). No read-only key; MCP balance checks and confirmation above 50 credits, a 30-day trash for workbooks and tables, but row, column and table deletes through the tools (12 of 20). Enrichment results include web-scraped and AI researcher text, with no prompt-injection guidance (3 of 15). Credit usage charts, transaction logs and per-request data logs (12 of 15). No security.txt, disclosure policy, bug bounty or certification found (0 of 20). 20+12+3+12+0 = 47.

Gojiberry 31

Credential line 18 to 20. Bearer API keys created in the app and shown once, now with a documented way to revoke them (deleting a key in Settings, API cuts its access at once). No scopes, and an organisation owner's key can act as any member through x-impersonate-user-id. The MCP server signs users in with OAuth and creates a key named after the client app (20 of 30). No read-only mode on the API. The MCP page says tools that change data are flagged so clients can ask first, and warns that send_unibox_message and the deletes can't be undone, but the server enforces nothing (3 of 20). Results carry LinkedIn profile text and inbound messages, with no prompt-injection guidance found (2 of 15). Campaign and agent logs and team activity through the API, a credit usage history in the app, no per-key call log found (6 of 15). No security.txt (404 on 4 November), disclosure policy, bounty or certification found. The privacy policy states encryption in transit and at rest, which is none of the programme line's items, so the line is 0 (was 3, a value with no programme evidence behind it). 20 + 3 + 2 + 6 + 0 = 31.

Schema & documentation Databar +13

Databar 84

Each API reference page embeds its OpenAPI 3.1 operation (spec version 0.1.2), with no downloadable file found (20 of 25). llms.txt with 79 API entries and Markdown twins (10). Descriptions say when to use each endpoint, such as headless runs against table runs, and what a delete leaves behind (16 of 20). Typed request bodies with required fields (12 of 15). An error envelope with stable codes and examples for 400, 406 and 422 (14 of 15). v1 paths and a dated changelog, though the API's own versioning isn't described (12 of 15). 20+10+16+12+14+12 = 84.

Gojiberry 71

Gojiberry moved its developer docs to dev.gojiberry.ai (Mintlify), and ext.gojiberry.ai/documentation now redirects there. Public OpenAPI 3.0.0 file, version 1.0, with 56 operations on 45 paths, linked from the docs' llms.txt under OpenAPI Specs (25). llms.txt 0 to 10. dev.gojiberry.ai serves llms.txt and every page as Markdown. 47 of 56 operations carry a description, and many say when to use them, such as the async enrichment routes for MCP clients and free search before paid reveal (15 of 20). 25 enums, 23 required lists and 50 required parameters, but search results are untyped objects and only four length bounds appear (9 of 15). 73 examples and curl samples. Error responses are a status code and a sentence, none of the 53 error responses has a body schema (8 of 15). A /v1 path and version 1.0, with no changelog found (4 of 15). 25 + 10 + 15 + 9 + 8 + 4 = 71.

Agent ergonomics Databar +11

Databar 66

69 MCP tools with no toolsets found (5), and REST row reads sized by per_page with has_next_page (10 of 25 for the two together). Pagination, row filters and a search parameter on enrichments (16 of 20). Errors name the fix, such as unknown columns with the available ones, 406 for credits, 409 for a stale internal_version and 410 for expired results (18 of 20). Upserts by key, version tokens on flow edits, cancel with refund, and MCP confirmation above 50 credits; MCP tool annotations not confirmed (13 of 20). Python SDK and a CLI, one language (9 of 15). 10+16+18+13+9 = 66.

Gojiberry 55

On the API, search returns masked rows until revealed, but a Contact has 75 fields and no field selection (14 of 25). page and limit on lists, eleven filters on contacts and fifteen on lead search (17 of 20). Status codes are documented per operation, with 402 for credits, 409 and 422. Bodies are {message, statusCode} and there's no error catalogue (11 of 20). No idempotency keys. Search is deterministic, a deleted contact can't be created again, and failed enrichments cost nothing. The MCP page says tools that change data are flagged, which we couldn't read from the server (7 of 20). Few required parameters and stated defaults, no official SDKs (6 of 15). 14 + 17 + 11 + 7 + 6 = 55.

Score by category

CategoryWeight this runDatabarGojiberryEdge
Reliability16%202870Gojiberry +42
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.28471Databar +13
Agent ergonomics13%16.26655Databar +11
Security & auth14%17.54731Databar +16
Payments & pricing10%12.53510Databar +25
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.87320Databar +53
Transparency & trust7%8.85561Gojiberry +6
Negative events≤1500
TotalD 53.8/100D 48.2/100

Facts side by side

FactDatabarGojiberry
KindHTTP APIHTTP API
VendorDatabar, Inc.SUPERFRUITS SAS (Gojiberry AI)
Hosted endpointhttps://mcp.databar.ai/mcphttps://ext.gojiberry.ai
TransportsHTTP, Streamable HTTP, SSE (legacy)HTTP, Streamable HTTP
AuthOAuth or keyOAuth or key
PricingPaidPaid
x402nono
LicenceProprietary, under Databar, Inc.'s terms of serviceProprietary service under Gojiberry's General Terms
Tools exposed69none
Read-only variant documentednono
llms.txtyesyes
Last release2026-10-30none
Terms last updated2026-10-01no date given
Privacy policy last updated2026-10-01no date given
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textyes
Terms restrict benchmarkingnot found in the textnot found in the text
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waiveryesnot found in the text

How the scores have moved

Verdicts

Databar

Databar puts 100+ enrichment providers, waterfalls and tables behind one REST API, a Python SDK, a CLI and a hosted MCP server with spending confirmation above 50 credits and stable error codes. API access needs a paid plan or trial, rate limits aren't published as numbers, and no status page or security programme was found.

Gojiberry

A public OpenAPI file with 56 operations linked from the developer docs' llms.txt, a free masked lead search and published limits of 100 requests a minute with Retry-After make the API easy to plan against. Keys have no scopes, an owner's key can act as any member, and writes can start LinkedIn and email outreach from connected accounts.

Before you call either

Databar

  1. Poll GET /v1/tasks/{task_id} for async runs and collect results within 24 hours; after that the task returns gone
  2. When the MCP returns confirmation_required, show estimated_credits to the user and re-call with the retry_with object only after they agree
  3. Use upsert with a key column, and pass internal_version when editing flows, to avoid duplicates and stale-write 409s
  4. Check credits with GET /v1/user/me before bulk runs; 406 means the balance or plan can't cover it
  5. Treat AI researcher and scraped outputs as untrusted text

Gojiberry

  1. Connect MCP clients to https://mcp.gojiberry.ai/mcp and sign in to Gojiberry. Custom clients must send a User-Agent header or get a 403
  2. Create resources in the order list, source agent, campaign. A source agent with no campaign only finds leads
  3. Adding a contact to a list enrols it in the linked campaign, which sends LinkedIn or email messages. Ask the user before list writes
  4. Search with POST /v1/directory/leads/search for free, then reveal only the ids you need at 0.5 credit each
  5. On a 429, wait for the Retry-After seconds before retrying. Limits are shared by every key and MCP connection of the account

Questions

Which is better for AI agents, Databar or Gojiberry?

Databar scores 53.8 (D) on agent readiness against Gojiberry's 48.2 (D), and leads in 5 of 7 scored categories. Gojiberry leads on reliability and transparency & trust.

Do Databar and Gojiberry need an API key?

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

Can an agent call Databar and Gojiberry without installing anything?

Yes. Databar has a hosted endpoint at https://mcp.databar.ai/mcp and Gojiberry at https://ext.gojiberry.ai.

Other comparisons with Databar or Gojiberry

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.