Do I need a Diffbot account?

For 15 platform tools, no. The other 7 require your own Diffbot key. All 22 accept your own key, and treg never meters BYOK calls.

What does a call cost?

Each tool on this page shows its rate; the cheapest is <$0.0001/result. The rate is the provider's own and treg.to adds no markup; it is billed per call from a prepaid balance. It is not Diffbot's subscription pricing, which is on their own site. New teams start with $1.00 of free credit.

How do I add Diffbot to Claude Code?

Run: claude mcp add --transport http treg https://treg.tracea.online/mcp. One MCP server carries Diffbot and the rest of the catalog.

Which frameworks does it work with?

Anything that speaks MCP (Claude Code, Claude Desktop, ChatGPT, Codex, Cursor, Grok) and anything that can make an HTTP request (LangChain, CrewAI, LlamaIndex, plain code).

Is this the official Diffbot MCP server?

No. treg.to serves Diffbot's real API through its own metered proxy: the request is the provider's own, the credential is injected server-side, and the answer is relayed verbatim. The official Diffbot channels are linked above.

Diffbot: 22 tools, platform or your own key

Q: How do I add Diffbot to Claude Code?

Run: claude mcp add --transport http treg https://treg.tracea.online/mcp. One MCP server carries Diffbot and the rest of the catalog.

Q: Which frameworks does it work with?

Anything that speaks MCP (Claude Code, Claude Desktop, ChatGPT, Codex, Cursor, Grok) and anything that can make an HTTP request (LangChain, CrewAI, LlamaIndex, plain code).

Q: Is this the official Diffbot MCP server?

No. treg.to serves Diffbot's real API through its own metered proxy: the request is the provider's own, the credential is injected server-side, and the answer is relayed verbatim. The official Diffbot channels are linked above.

The shelf

All 22 Diffbot tools

Open web & SEO data · 17 tools

Check plan, usage and credit balance · BYOK only · account management · live-verified · No treg charge; upstream free
Retrieves account plan tier, monthly credit allowance, token status, child tokens and a daily usage breakdown by API diffbot.x.account
account/usage lookups are not metered
Manage a crawl job · BYOK only · account management · live-verified · No treg charge; upstream
One endpoint for status, pause/resume, restart and delete of one or more crawl jobs diffbot.x.crawl-manage
Auto-detect a page type and extract it by URL · Platform + BYOK · live-verified · $0.0012/call
Automatically identifies the type of page being processed and routes it through the relevant Diffbot API diffbot.x.extract-analyze
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract article text, author and date from a URL · Platform + BYOK · live-verified · $0.0012/call
Extracts clean article text, author, date, images, tags, sentiment and NLP facts from a news or blog page diffbot.x.extract-article
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract forum posts and comments from a thread URL · Platform + BYOK · live-verified · $0.0012/call
Extracts individual posts, authors, dates and sentiment from a forum thread or comment section diffbot.x.extract-discussion
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract event name, date and venue from a page URL · Platform + BYOK · live-verified · $0.0012/call
Extracts event name, date, venue, price and organizer from an event listing page diffbot.x.extract-event
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract a page's images by URL, with dimensions · Platform + BYOK · live-verified · $0.0012/call
Extracts the primary and secondary images from a page, with dimensions and metadata diffbot.x.extract-image
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract a job posting page by URL — title, salary · Platform + BYOK · live-verified · $0.0012/call
Extracts job title, employer, location, salary and description from a job posting page diffbot.x.extract-job
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract linked items from an index page by URL · Platform + BYOK · live-verified · $0.0012/call
Extracts the list of linked items (with titles, dates, images) from an index or category page diffbot.x.extract-list
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract product name, price and reviews from a URL · Platform + BYOK · live-verified · $0.0012/call
Extracts product name, price, SKU, images, specs and reviews from an e-commerce page diffbot.x.extract-product
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract video title, duration and embed URL from a page · Platform + BYOK · live-verified · $0.0012/call
Extracts title, embed/player URL, duration, thumbnail and metadata from a video page diffbot.x.extract-video
1 credit per page (2 with useProxy=default) OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The 2-credit useProxy=default rate was not exercised and stays documented.
Extract entities, sentiment and facts from text · Platform + BYOK · live-verified · <$0.0001/result
Runs a pre-trained classifier, named-entity recognizer, sentence tokenizer and sentiment analyzer over submitted text or a URL diffbot.x.nl-process
1 credit per 10,000 characters of input (document cap 100,000 chars; request cap 1,000,000 chars) — scales with input length, no single fixed per-call figure to record
Web search shaped for an LLM context window · BYOK only · live-verified · No treg charge; upstream
Search the web and return results shaped for feeding into an LLM context window diffbot.x.web-search
Start a bulk extract job from a list of URLs · BYOK only · account management · No treg charge; upstream own account
Sends a large, explicit list of URLs through an Extract API asynchronously and compiles the results into one collection — unlike Crawl, it does not discover links diffbot.x.bulk-create
no flat job charge — each URL processed bills at its apiUrl's per-page rate (1-2 credits); no single per-call figure to record
Download a bulk extract job's results as JSON or CSV · BYOK only · No treg charge; upstream
Downloads the extracted results of a bulk job as JSON or CSV, or a diagnostic URL report diffbot.x.bulk-data
Start a site crawl job from seed URLs · BYOK only · account management · No treg charge; upstream own account
Spiders a site for links and pages, running each discovered page through an Extract API and compiling results into a searchable collection diffbot.x.crawl-create
no flat job charge — each page the crawl sends through apiUrl bills at that Extract API's per-page rate (1-2 credits); no single per-call figure to record
Download a crawl job's results as JSON or CSV · BYOK only · No treg charge; upstream
Downloads the extracted results of a crawl job as JSON or CSV, or a diagnostic URL report diffbot.x.crawl-data

Company data · 3 tools

Bulk-enrich companies or people (async job) · Platform + BYOK · live-verified · $0.0299/result
Enrich many companies or people in one asynchronous job instead of one call each diffbot.companies.bulk_enrich
25 credits per record in the batch (100 with refresh) — the same rate as a single /enhance call, so batching saves round trips, not credits. OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The meter counts ENTITIES RETURNED, not calls, so a lookup answering with one record costs 25 and a multi-record response scales; every observed call returned exactly one.
Enrich a company by website, name or ticker · Platform + BYOK · live-verified · $0.0299/call
Enrich a company from website, name or ticker — industry, revenue, headcount, rivals diffbot.companies.enrich
25 credits per enhanced record, or 100 credits with refresh=true. `size` multiplies this. On the $299/mo 250k-credit plan that is roughly $0.03 per record. OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The meter counts ENTITIES RETURNED, not calls, so a lookup answering with one record costs 25 and a multi-record response scales; every observed call returned exactly one.
Search companies by DQL filter — industry, revenue · Platform + BYOK · live-verified · $0.0299/result
Search the Diffbot Knowledge Graph for companies by DQL filter (industry, revenue) diffbot.companies.search
25 credits per entity record returned (100 per facet-query result), so `size` is the price. Billed per result, not per query. OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The meter counts ENTITIES RETURNED, not calls, so a lookup answering with one record costs 25 and a multi-record response scales; every observed call returned exactly one.

People & contact data · 2 tools

Enrich a person by name, email or employer · Platform + BYOK · live-verified · $0.0299/call
Enrich a person from name, email or employer — role, employment history, profiles diffbot.people.enrich
25 credits per enhanced record, or 100 with refresh=true — same rate as company enrichment. OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The meter counts ENTITIES RETURNED, not calls, so a lookup answering with one record costs 25 and a multi-record response scales; every observed call returned exactly one.
Search people by DQL filter — role, employer, skills · Platform + BYOK · live-verified · $0.0299/result
Search the Diffbot Knowledge Graph for people by DQL filter (role, employer, skills) diffbot.people.search
25 credits per entity record returned (100 per facet-query result); `size` is the price. OBSERVED 2026-07-31 — GET /v4/account's daily usage rows reconcile exactly to credits = 25*entities + 1*extractions + 1*nlp on two independent days — 2026-07-28 billed 136 for 5 entities + 10 extractions + 1 nlp, and 2026-07-30 billed 60 for 2 + 9 + 1, which is the whole charge with nothing left over. The meter counts ENTITIES RETURNED, not calls, so a lookup answering with one record costs 25 and a multi-record response scales; every observed call returned exactly one.

Reliability badges come from live traffic through treg.to. Live-verified means a successful provider check; it does not verify every request variant.