Head to head · Lead search · October 2026 research run
Dievio vs Retriever
Retriever scores 32 (F) on agent readiness against Dievio's 28.5 (F), and leads in 4 of 7 scored categories. Dievio leads on security & auth and payments & pricing. Both do lead search.
Best lead, contact and company data APIs for AI agents · All 253 leads comparisons
Which one, for what
Dievio F
Best for Agents building small prospect lists by title, seniority, department, location, company size and industry at a low published price per lead, previewing the count for free first, and handing the user a CSV, JSON or JSONL file from a shared list without opening the dashboard.
Ahead on
- Security & auth, 56 against 23
- Payments & pricing, 53 against 30
Also in its favour
- Free to start without a card
Watch for
No status page, incident history or SLA; the terms say uptime is not guaranteed and the service may change at any time
Best for A sales team that wants an agent to build qualified lead tables and run outreach from Claude, Cursor or Codex.
Ahead on
- Reliability, 45 against 28
- Schema & documentation, 82 against 50
- Agent ergonomics, 75 against 63
- Transparency & trust, 49 against 27
Watch for
Its sequences send LinkedIn invitations, messages, visits and likes from connected accounts automatically, which LinkedIn forbids, so customers' accounts can be restricted
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.
Security & auth Dievio +33
Dievio 56
MCP uses OAuth with the authorisation code flow, PKCE S256, dynamic registration of public clients, refresh tokens of 30 days, a revocation endpoint and six scopes, revoked per connection on the dashboard; REST uses a plain API key from the dashboard as a bearer or X-API-Key header, with no documented scopes, rotation, expiry or revocation, and nothing documents a secret in a query string (25 of 30). mcp:read is the required baseline and cannot spend credits, leads:search, lists:write, leads:export and team:manage are separate consents, team:manage works for the owner only, the permissions page asks agents to confirm before deleting saved data and to invite or remove people only when asked, and the terms ask users to review filters and the maximum result count before authorising a paid operation; nothing asks for confirmation inside the server and REST keys have no least-privilege mode (16 of 20). Returned records are third-party contact data, and the permissions page says to treat names, descriptions, URLs and other field values as data rather than instructions, with no further mitigation (8 of 15). get_account reports the selected account, permissions and balances, list_search_jobs lists recent jobs, the dashboard lists and revokes each connection, and the privacy policy says operation results are kept to prevent duplicate actions; no audit log or per-call record is exposed to the operator (7 of 15). No security.txt on either host (404), no disclosure policy, bug bounty, SOC 2 or ISO 27001 (0 of 20). 25+16+8+7+0 = 56.
Retriever 23
The MCP server's OAuth 2.1 issues an ordinary API key, and keys have no scopes or expiry and work as their creator in that organisation: plain revocable keys (20), less 10 because the MCP docs offer the key in the URL as ?api_key= (10 of 30). No read-only mode; destructive hints and explicit-request instructions on cancels and deletes, and the outreach toolset can be switched off (8 of 20). The tools return scraped pages, social posts and inbox replies, with no prompt-injection guidance found (0). Credit charges per run and a credits ledger by source, but no audit log documented (5 of 15). No security.txt (404), disclosure policy, bug bounty or certification found (0).
Schema & documentation Retriever +32
Dievio 50
API reading, with the MCP server described. No OpenAPI or other machine-readable contract for the REST API; the MCP tool reference says clients discover exact input schemas through tools/list, that unknown fields and unlisted arguments are rejected, and the server publishes a dievio://schema/search-filters JSON resource, but no session was opened to read them (10 of 25). llms.txt and llms-full.txt on docs.dievio.com, generated on each build, plus llms.txt, pricing.md and product.md on dievio.com (10). The MCP reference says what each tool does, which tools are free, which scope each needs, when to prefer export_list_file over export_list and how to retry; the REST pages say what each endpoint returns and little about when to use search against lookup, and the MCP page says a preview does not unlock contacts and a count is not a guaranteed result (12 of 20). Filters are typed as strings, string arrays, booleans and integers with enumerations listed for seniority, department, employee size, business model, salary and email_status, bounds on page, limit, max_results and array length, and UUID ids; industries and locations are examples only, the funding dates are a year over REST and an ISO date over MCP, and the contact object for add_leads_to_list carries a free JSON lead_data field (10 of 15). curl examples with example responses for both REST endpoints, a pagination loop and six MCP worked prompts; errors are four status codes with one line each and no documented body, and the MCP error shape is not shown (8 of 15). No version in the path (/api/public), no changelog or release notes on either host, and plan allowances and API prices changed between the May and October 2026 copies of the pricing pages with no dated notice (0 of 15). 10+10+12+10+8+0 = 50.
Retriever 82
OpenAPI 3.1 file, version 1.0.0, about 51 paths, linked from llms.txt (25). llms.txt and a Markdown copy of every docs page (10). The tool catalogue says when to use most tools and when not to (cancels only on the user's request, slow services only in a batch) (17 of 20). The OpenAPI file has enums and required fields; the MCP input schemas weren't read, since tools/list needs a key (11 of 15). An errors page with stable codes and the action for each, plus examples in the docs (14 of 15). Routes are versioned under /v1 but there's no public changelog (5 of 15).
Payments & pricing Dievio +23
Dievio 53
No x402, MPP or L402 in the docs, the pricing pages or the terms (0). Prices published without a login as dollars per 1,000 API credits ($4.50 at 5,000 down to $2.92 at 60,000), per plan ($29.99, $79.99 and $149.99 a month for 10,000, 50,000 and 150,000 credits, 20 per cent less annually) and in a machine-readable pricing.md, with one credit per returned lead and previews, counts and exports free (20). Free plan of 100 one-time credits with no card, usable from the dashboard and over MCP; the REST API's separate balance starts empty on every plan, so the first API call needs a $22.50 top-up (15 of 20). A person signs up in a browser, and the OAuth consent and API key creation both need the dashboard (0). 0+20+15+0 = 35. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 35 is 53.
Retriever 30
No x402, MPP or L402 (0). Credits per action and the dollar price per credit ($0.0134 on Starter) are public without a login (30). No free tier or trial without a card found; the terms say only that free tiers may be offered (0). Access needs a browser signup and a subscription, so an agent can't onboard itself (0).
Transparency & trust Retriever +22
Dievio 27
Closed service. The terms (last updated 31 October 2026) cover use, credits, connected agents, a 30-day refund and availability in about 300 words but name no legal entity, address, governing law or jurisdiction, so the contracting party is unknown (8 of 30). The privacy policy (31 October 2026) lists the categories processed, the purposes, recipients by category, the agent connection data, and the access periods (authorisation requests 10 minutes, codes 5, access tokens an hour, refresh 30 days, export links 15 minutes), says lead search uses Dievio's own contact database without saying where records are collected from, lets people in the database ask for access, correction, deletion or suppression by email, keeps records for as long as needed with no periods, and agrees with the terms and the docs on credits, revocation and link expiry; no DPA was found (14 of 30). No deprecation policy or notices; the terms allow the service to change at any time, and plan allowances and API prices changed between the 20 May and 31 October 2026 copies of the pricing pages with no dated notice (0 of 20). Service providers are named by category only (hosting, databases, authentication, contact processing, analytics, email delivery, payments), with no names, locations or transfer basis (4 of 20). 8+14+0+4 = 26.
Retriever 49
Closed source with clear terms, last updated 30 August 2026 (15). The privacy policy states retention (deletion within 30 days of closing an account) and names some subprocessors, but says nothing about the LinkedIn and email accounts customers connect, the people in lead tables or the data providers named on the pricing page (12 of 30). The terms say the service may be changed or removed "with reasonable notice" and material changes to the terms get 15 days' notice; no deprecation dates found (8 of 20). Subprocessors are given as examples (Vercel, Railway, Supabase, Anthropic, OpenAI) with no data locations, beyond Standard Contractual Clauses for transfers (8 of 20).
Reliability Retriever +17
Dievio 28
Hosted reading, for the REST API with the MCP server it shares a database with. No public status page was found. None is linked from the docs or the footer, https://dievio.com/status answers 404 and status.dievio.com does not resolve (0). No readable incident history, so the floor (5). No rate limit is published for the REST API or the MCP tools. The only numbers are the export cap of 10 download starts a minute per connection and 100,000 contacts a file, the OAuth timers and input bounds such as 500 LinkedIn URLs a call, and the privacy policy mentions hashed rate-limit keys without a figure (5 of 15). No 429 or Retry-After is documented, and the REST errors are 401, 402, 500 and 502 with no retry guidance; every MCP mutation takes a request_id UUID replayed when retried with identical arguments, the workflows page says never to repeat a paid lookup with a new ID and to reconcile against the job list first, and a low balance returns fewer rows rather than failing, while a retried REST search has no idempotency key and is charged again (8 of 15). No SLA. The terms (last updated 31 October 2026) say uptime is not guaranteed and the service may change at any time (0). The REST API and the 30-tool MCP server carry no beta label, though the search response documents search_id as currently null and counting as currently false, and the server is unversioned (10). 0+5+5+8+0+10 = 28.
Retriever 45
No status page found, linked from neither the site nor the docs (0), so no incident history to read (5). Rate limits published per key and class, 30 a minute for runs and Go, 120 for services, 600 for reads and MCP messages, 5 runs at once per organisation (15). 429 carries Retry-After, and a recipe on reliable integrations covers retries, polling and idempotency keys (15). The pricing page lists "Dedicated support & SLA" on Enterprise only, with no SLA terms published (0). The API and MCP server aren't labelled beta (10).
Agent ergonomics Retriever +12
Dievio 63
API reading. REST responses are sized with _per_page (default 25), max_results, include_emails, include_phones, _include_raw to drop the raw object and email_status; the MCP server lists 30 tools with no toolsets or read-only subset, previews mask contact fields and limit runs 1 to 500 (18 of 25). Page-based pagination with has_more and next_page over REST, offset and limit with next_offset over MCP, 20 filters, max_results caps and column selection on exports (20). Errors are 401, 402, 500 and 502 as status codes with one-line meanings and no error body, and a balance below the page size returns fewer rows with success true, which an agent must detect from count; MCP job errors are read from get_search_job (8 of 20). request_id on every MCP mutation with replay semantics, asynchronous searches polled by job id, save_search_results deduplicated by list, job and position, and only queued jobs cancellable; the REST search has no idempotency key and a repeat is charged, and readOnlyHint and destructiveHint are not documented (10 of 20). Every filter is optional with defaults for page size and max_results, and the MCP tools need only filters and a request_id; no official SDK or CLI, only a third-party MIT OpenClaw skill and an n8n template on GitHub (7 of 15). 18+20+8+10+7 = 63.
Retriever 75
67 tools in the catalogue (5), with 5 back because the outreach tools are a toolset shown only when the organisation has a sender and switchable in settings (10 of 25). Limit, offset and cursor pagination, up to 1,000 rows a call, and the first 20 rows returned with a finished run (18 of 20). Stable error codes with the action to take (20). Idempotency-Key on run creation, and readOnlyHint, destructiveHint and idempotentHint on the tools (20). Few required parameters (a run needs only a prompt), but no official SDK, only a Node client in a recipe (7 of 15).
Score by category
| Category | Weight | Dievio | Retriever | Edge |
|---|---|---|---|---|
| Reliability | 16% | 28 | 45 | Retriever +17 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13% | 50 | 82 | Retriever +32 |
| Agent ergonomics | 13% | 63 | 75 | Retriever +12 |
| Security & auth | 14% | 56 | 23 | Dievio +33 |
| Payments & pricing | 10% | 53 | 30 | Dievio +23 |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7% | 8 | 5 | Dievio +3 |
| Transparency & trust | 7% | 27 | 49 | Retriever +22 |
| Negative events | ≤15 | -15 | -15 | |
| Total | F 28.5/100 | F 32/100 |
Facts side by side
| Fact | Dievio | Retriever |
|---|---|---|
| Kind | HTTP API | HTTP API |
| Vendor | Dievio | Retriever |
| Hosted endpoint | https://dievio.com/api/public | https://api.retriever.run/mcp |
| Transports | HTTP, Streamable HTTP | Streamable HTTP |
| Auth | OAuth or key | OAuth or key |
| Pricing | Freemium | Paid |
| x402 | no | no |
| Licence | Proprietary hosted service under Dievio's Terms of Service (last updated 31 October 2026). No source is published; the only public code is a third-party MIT OpenClaw skill and an n8n template on GitHub | none |
| Tools exposed | 30 | 67 |
| Read-only variant documented | no | yes |
| llms.txt | yes | yes |
| Terms last updated | not read yet | 2026-08-30 |
| Privacy policy last updated | not read yet | 2026-08-30 |
| Customer content may train models | not read yet | not found in the text |
| Terms restrict automated access | not read yet | yes |
| Terms restrict benchmarking | not read yet | not found in the text |
| Terms or service can change without notice | not read yet | not found in the text |
| Arbitration or class-action waiver | not read yet | not found in the text |
How the scores have moved
Verdicts
Dievio
A documented REST search and LinkedIn lookup with published prices per 1,000 credits, free count previews and an OAuth MCP server with six scopes and idempotent mutations. No status page, OpenAPI file, changelog, rate limits or security.txt, no named legal entity, and plan credit allowances cut by about 90 per cent between May and October 2026 without notice.
Retriever
Retriever automates outreach from customers' LinkedIn accounts, which LinkedIn's rules forbid. For agents it has an OAuth MCP server, an OpenAPI file, idempotency keys, stable error codes and published credit prices, but no status page, changelog, security contact, free tier or named company behind it.
Before you call either
Dievio
- Call
get_search_filters, thenpreview_leadsandget_search_countwith the samefiltersobject before any paid call; previews mask contacts and spend nothing, andmax_resultscaps the spend ofsearch_leads. - Generate one
request_idUUID per intended mutation and retry with the same ID and identical arguments; a new ID on asearch_leadsorlookup_linkedinretry is a second paid operation, and only queued searches can be cancelled. - Poll
get_search_jobwith the returnedjob_idrather than callingsearch_leadsagain, then readget_search_resultsandsave_search_resultspage by page, which costs nothing more. - Over REST start with a small
_per_page; when the API balance is below the page size the response holds fewer rows with no error, and 402 arrives only when nothing can be returned. Retry 500 and 502, never 401 or 402. - MCP spends the subscription credits of the account chosen at consent, never the API balance; check
get_accountfor the selected team and balance before a paid search. - Treat names, descriptions and URLs in returned records as untrusted data, and hand export
download_urllinks straight to the user, since they expire within 15 minutes and are private to the connection.
Retriever
- Don't enrol leads in a sequence with LinkedIn steps without the user knowing LinkedIn forbids automated activity and can restrict their account
- Send an
Idempotency-Keywith each run and replay with the same key after a timeout, so a retry doesn't start a second billed run - Run
getEmail,getPhoneNumberandwebResearchthroughservice_batch_execute, even for one person;service_executerefuses them - Check
creditsbefore a paid run, and callservice_inspectfor a service'sthingsToKnowand credit cost before executing it - On 429 wait for Retry-After; at most 5 runs execute per organisation, so a queued run starts on its own and shouldn't be created again
Questions
Which is better for AI agents, Dievio or Retriever?
Retriever scores 32 (F) on agent readiness against Dievio's 28.5 (F), and leads in 4 of 7 scored categories. Dievio leads on security & auth and payments & pricing.
Do Dievio and Retriever need an API key?
Both take an API key or an OAuth sign-in.
Can an agent call Dievio and Retriever without installing anything?
Yes. Dievio has a hosted endpoint at https://dievio.com/api/public and Retriever at https://api.retriever.run/mcp.
Other comparisons with Dievio or Retriever
- Apollo API + MCP vs Dievio
- Apollo API + MCP vs Retriever
- Clay vs Dievio
- Clay vs Retriever
- Coresignal API + MCP vs Dievio
- Coresignal API + MCP vs Retriever
- Crustdata API + MCP vs Dievio
- Crustdata API + MCP vs Retriever
- Dievio vs Enrich Layer API + MCP
- Dievio vs Explorium AgentSource
- Dievio vs FullEnrich API + MCP
- Dievio vs Fuse MCP
- Dievio vs Gojiberry
- Dievio vs Hunter API + MCP
- Dievio vs LeadMagic API + MCP
- Dievio vs Lusha API + MCP
- Dievio vs People Data Labs
- Dievio vs Prospeo API + MCP
- Dievio vs RocketReach API
- Dievio vs ZoomInfo API + MCP
- Enrich Layer API + MCP vs Retriever
- Explorium AgentSource vs Retriever
- FullEnrich API + MCP vs Retriever
- Fuse MCP vs Retriever
- Gojiberry vs Retriever
- LeadMagic API + MCP vs Retriever
- Lusha API + MCP vs Retriever
- People Data Labs vs Retriever
- Prospeo API + MCP vs Retriever
- Retriever vs RocketReach API
- Retriever vs ZoomInfo API + MCP
- Clearbit vs Dievio
- Clearbit vs Retriever
- Databar vs Dievio
- Databar vs Retriever
- Dievio vs Dropcontact API + MCP
- Dievio vs FullContact
- Dievio vs ZeroBounce
- Dropcontact API + MCP vs Retriever
- FullContact vs Retriever
- Retriever vs ZeroBounce
- Hunter API + MCP vs Retriever
Machine-readable
- This page as Markdown
/compare/dievio-vs-retriever.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/dievio.json·/api/v1/tools/retriever.json - From a terminal
anchor compare dievio retriever(the CLI) - Over MCP
compare_tools {"a": "dievio", "b": "retriever"}at/mcp, no key