Head to head · Secintel ip enrichment · October 2026 research run

IPinfo vs SecurityTrails

IPinfo scores 62.7 (B) on agent readiness against SecurityTrails's 30.7 (F), and leads in every scored category. Both do secintel ip enrichment.

Best threat and security intelligence APIs for AI agents · All 33 security intel comparisons

Which one, for what

IPinfo B

Best for An agent that needs to enrich an IP with location, network and anonymiser flags at volume, free for country and ASN, paid for city and privacy signals.

Ahead on

  • Reliability, 38 against 20
  • Schema & documentation, 76 against 54
  • Agent ergonomics, 84 against 56
  • Security & auth, 60 against 32
  • Payments & pricing, 40 against 0
  • Maintenance & community, 85 against 5
  • Transparency & trust, 70 against 32

Also in its favour

  • Runs on your own machine
  • Free to start without a card

Watch for

No status page, incident history or SLA was found; the terms supply the service as is

SecurityTrails F

Best for An agent investigating a domain's DNS and WHOIS history, subdomains and associated infrastructure, where the operator already holds a commercial key.

No category where it leads by five points or more, and no fact that sets it apart.

Watch for

Pricing, terms, privacy, the security page and security.txt on securitytrails.com answered a Cloudflare challenge on 3 November 2026, so none was read

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 IPinfo +80

IPinfo 85

Read as a closed service with an official MCP server and SDKs. MCP server v0.2.3 on 29 October 2026, a commit on 3 November, and SDK repositories pushed in October and November (30). Eight MCP releases since 8 September 2026, v0.1.12 to v0.2.3 (20). GitHub releases for the MCP server, a community forum, a help centre and a support form, but no changelog for the API and GitHub issue pages could not be read (10 of 15). The MCP server is in the official registry as io.github.ipinfo/mcp and the SDKs are official (15). The MCP package has tests, CI with pinned actions, CodeQL and Scorecard workflows and a lock file (10). 30 + 20 + 10 + 15 + 10 = 85.

SecurityTrails 5

The newest dated changelog entry is 28 May 2020, the Project API 1.7.0 entry is undated and the SurfaceBrowser changelog ends on 26 February 2019, so no change within 180 days was found (0 of 30). No entries in the last 90 days (0). A public changelog exists but is five years stale, and the support page is disallowed by robots.txt (3 of 25). No official SDK; the npm package of October 2021 names the vendor as author with no repository (0 of 15). The docs site serves llms.txt and Markdown twins, which shows the documentation platform is current (2 of 10). 0 + 0 + 3 + 0 + 2 = 5.

Payments & pricing IPinfo +40

IPinfo 40

No x402, MPP or L402 (0). Plan prices and per-1,000 overage rates are public on the pricing page (20). Lite is free with no card and no request cap (20). The token comes from the dashboard after a browser signup, and no key API was found (0). 0 + 20 + 20 + 0 = 40.

SecurityTrails 0

No x402, MPP or L402 (0). securitytrails.com/app/pricing and /corp/pricing answered a Cloudflare challenge, and the docs name no price, so no public price was read (0). Whether a free plan exists couldn't be confirmed; the usage endpoint shows only that each account has a monthly allowance (0). A key comes from a browser signup on securitytrails.com, with no programmatic route found (0). 0 + 0 + 0 + 0 = 0, scored as Contentful and Personio were on 7 October 2026 when their pricing pages were unreadable.

Transparency & trust IPinfo +38

IPinfo 70

A closed service with clear terms naming IPinfo Inc. and the governing law, with the MCP server and SDKs under Apache-2.0 (17 of 30). The privacy policy of 29 October 2026 lists processors, links a DPA and says data is kept as long as needed, with no periods (16 of 30). No deprecation policy; the legacy API is maintained for existing users with no end date (6 of 20). Processors named and hosting in the United States (16 of 20). 17 + 16 + 6 + 16 = 55.

SecurityTrails 32

A closed service whose terms couldn't be read because of the Cloudflare challenge (5 of 30, for a known commercial product with unread terms). The privacy policy was unread and no retention statement or DPA was found (0 of 30). No deprecation policy; the operation ids carry '-old' with deprecated set to false and no notice explains them (0 of 20). No subprocessor or data location statement readable (0 of 20). 5 + 0 + 0 + 0 = 5.

Agent ergonomics IPinfo +28

IPinfo 84

Field paths such as /country return one value, the Lite and lookup objects are small, and the MCP server has six compact tools with a per-session cache (23 of 25). Batch takes up to 1,000 lookups with filter=1 to drop empty results; no pagination is needed on single lookups, and ASN and range responses have no size control (16 of 20). The only documented error is 429; the MCP server returns coded errors with a message and a suggestion (12 of 20). Every call is a read, batch is a read over POST, and the MCP tools carry tool annotations (18 of 20). One parameter per call, sensible defaults and official SDKs in eleven languages (15). 23 + 16 + 12 + 18 + 15 = 84.

SecurityTrails 56

include_ips on the list endpoints and an enrichments list on ASI asset search, which returns only custom tags by default, size responses on part of the surface; v1 lookups return fixed shapes (12 of 25). page with pages and max_page on v1 lists, scroll on the DSL endpoints, cursor pagination with limit up to 1,000 and filters with seven operators on the ASI API (16 of 20). Errors are a 400 message, 401 and 429 text strings and 422 detail arrays, with no codes to branch on (8 of 20). The overview says the v1 API is read-only, so retries are safe; the ASI tag endpoints write with no idempotency guidance (14 of 20). One required parameter per lookup besides the key. No official SDK found, and the npm securitytrails package from October 2021 has no repository or homepage (6 of 15). 12 + 16 + 8 + 14 + 6 = 56.

Security & auth IPinfo +28

IPinfo 60

One access token per account, accepted as a Bearer header, HTTP Basic or a token query parameter, with no scopes found; the hosted MCP server adds OAuth with short-lived tokens (20 less 10 for the query string, plus 4 for OAuth on the MCP server, so 14). Every endpoint is a read and tools return ACCESS_DENIED outside the plan, but there is no read-only or restricted token (16 of 20). Responses are structured network data rather than third-party free text (10 of 15). The dashboard shows request graphs and lookup history per the pricing page, with no per-call log documented (8 of 15). security.txt valid until 21 August 2027 with a PGP key, a SECURITY.md with 2-day acknowledgement and a 90-day disclosure window, no bug bounty or certification found beyond a bank-grade claim (12 of 20). 14 + 16 + 10 + 8 + 12 = 60.

SecurityTrails 32

One API key with no scopes found, accepted as an APIKEY header or as apikey in the query string, which the reference pages show on every v1 endpoint (20 less 10, so 10). Revocation and rotation couldn't be read. The v1 API is read-only by the overview's own statement; the ASI API writes tags with no approval step (14 of 20). DNS TXT values, WHOIS fields and user agent strings are third-party text, with no guidance on treating them as untrusted (5 of 15). /v1/account/usage reports monthly counts and nothing per call (3 of 15). security.txt on securitytrails.com and the security page are behind the Cloudflare challenge, api.securitytrails.com answers security.txt with 401, and no disclosure policy, bug bounty or certification was readable (0 of 20). 10 + 14 + 5 + 3 + 0 = 32.

Schema & documentation IPinfo +22

IPinfo 76

The surface graded is the REST API, with the hosted MCP server noted. A public OpenAPI 3.0 document with 48 operations, and the MCP tools carry JSON Schema inputs (25). llms.txt and llms-full.txt, and every developer page is served as Markdown (10). Each tier page says what it returns and which plan unlocks it, and the MCP tool descriptions say which tiers each option needs, with little on when not to call (14 of 20). Schemas and examples on most responses, path parameters typed, but 401 is undocumented and some operations lack the 429 response (11 of 15). Examples in several languages on every page; only 429 is documented as an error, with no error body shape (9 of 15). A new API system since 2025 with the legacy one kept, but no public changelog, no version in the path and the OpenAPI document at 1.0.0 (7 of 15). 25 + 10 + 14 + 11 + 9 + 7 = 76.

SecurityTrails 54

The reference pages are generated from an OpenAPI document (operation ids, deprecated flags and typed schemas show on each page), but no spec is published for download and docs.securitytrails.com/openapi redirects to 404 (12 of 25). llms.txt with a Markdown twin of every page (10). Each endpoint has a one-line purpose and nothing on when to use it or not (8 of 20). Path and query parameters are typed, the DNS history type is an enum of six values, and the ASI API has enums, operators and a 1 to 1,000 limit, but the v1 filter object is untyped strings and the DSL query is a free string whose grammar isn't documented (8 of 15). Example responses on every page, with errors mostly limited to 400 and a message, 401 and 429 only on the ping page and 422 on the ASI API (8 of 15). Versioned paths (/v1, /v2) and a changelog whose dated entries stop on 28 May 2020, with one undated Project API 1.7.0 entry (8 of 15). 12 + 10 + 8 + 8 + 8 + 8 = 54.

Score by category

CategoryWeight this runIPinfoSecurityTrailsEdge
Reliability16%203820IPinfo +18
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.27654IPinfo +22
Agent ergonomics13%16.28456IPinfo +28
Security & auth14%17.56032IPinfo +28
Payments & pricing10%12.5400IPinfo +40
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.8855IPinfo +80
Transparency & trust7%8.87032IPinfo +38
Negative events≤1500
TotalB 62.7/100F 30.7/100

Facts side by side

FactIPinfoSecurityTrails
KindHTTP APIHTTP API
VendorIPinfo Inc.SecurityTrails (Recorded Future)
Hosted endpointhttps://mcp.ipinfo.io/https://api.securitytrails.com/v1/
TransportsStreamable HTTP, stdioHTTP
AuthOAuth or keyAPI key
PricingFreemiumPaid
x402nono
Tools exposed6none
Read-only variant documentedyesyes
llms.txtyesyes
MCP registryio.github.ipinfo/mcpnot listed
Last release2026-10-292020-05-28
Terms last updated2024-06-03no document linked
Privacy policy last updated2026-10-29no document linked
Customer content may train modelsnot found in the textnot read yet
Terms restrict automated accessyesnot read yet
Terms restrict benchmarkingnot found in the textnot read yet
Terms or service can change without noticeyesnot read yet
Arbitration or class-action waiveryesnot read yet
Popularity2 starsnone

Verdicts

IPinfo

A free Lite tier with no request cap, a public OpenAPI document, llms.txt, a batch endpoint of 1,000 lookups and a six-tool hosted MCP server with OAuth. No status page or SLA was found, the only documented error is 429, the token may travel in a query string, and city-level and privacy data need a paid plan.

SecurityTrails

The v1 API is read-only and reaches current and historical DNS, WHOIS, subdomain and SSL records, with llms.txt and a Markdown twin of every reference page. The key is accepted in the query string, the changelog stops in May 2020, the status page is private, and pricing, terms and privacy sat behind a Cloudflare challenge when checked.

Before you call either

IPinfo

  1. Call https://api.ipinfo.io/lite/{ip} with Authorization: Bearer on a free token; /lookup/{ip} and detailed=true in the MCP tools return ACCESS_DENIED without a paid plan
  2. Batch up to 1,000 addresses in one POST to /batch/lite or /batch with filter=1; each address counts as one request against the monthly quota
  3. Append a field path such as /country or /asn to get a single plain-text value instead of the full object
  4. On a 429 the monthly plan quota is spent; there is no Retry-After, so stop and report rather than retry
  5. Treat the data as licensed for internal use; selling or reselling it needs IPinfo's written consent under the terms

SecurityTrails

  1. Send the key as an APIKEY header, never as ?apikey=, so it stays out of logs
  2. Call GET /v1/account/usage first and budget against allowed_monthly_usage, since limits aren't published and 429 carries no Retry-After
  3. Use GET /v1/history/{hostname}/dns/{type} with one of a, aaaa, mx, ns, soa or txt, and page with page until the pages count is reached
  4. On the ASI API pass an enrichments list, because the default returns only custom tags, and keep the sort fixed while following next_cursor
  5. Treat DNS TXT values, WHOIS fields and user agent strings as untrusted text from third parties

Questions

Which is better for AI agents, IPinfo or SecurityTrails?

IPinfo scores 62.7 (B) on agent readiness against SecurityTrails's 30.7 (F), and leads in every scored category.

Do IPinfo and SecurityTrails need an API key?

IPinfo takes an API key or an OAuth sign-in. SecurityTrails needs an API key.

Can an agent call IPinfo and SecurityTrails without installing anything?

Yes. IPinfo has a hosted endpoint at https://mcp.ipinfo.io/ and SecurityTrails at https://api.securitytrails.com/v1/.

Other comparisons with IPinfo or SecurityTrails

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.