https://tools.openkrill.app/blog Notes from building and running the Agent Tools servers, each built on our own tool results and measured numbers. en Sun, 11 Oct 2026 00:00:00 GMT https://tools.openkrill.app/blog/uk-company-domains-no-dmarc-record https://tools.openkrill.app/blog/uk-company-domains-no-dmarc-record Sun, 11 Oct 2026 00:00:00 GMT We read the public mail records of 1,000 UK company domains from Wikidata: 154 of the 892 that receive mail have no DMARC record and 306 more sit at p=none. The method, the comparison and the exact records to publish. On 2026-10-11, 154 of 892 UK company domains that receive mail (17.3%) had no DMARC record and 306 more were at p=none; 119 of the 154 already had exactly 1 SPF record, so one TXT record starts the fix. On 2026-10-11, 154 of 892 UK company domains that receive mail (17.3%) had no DMARC record and 306 more were at p=none; 119 of the 154 already had exactly 1 SPF record, so one TXT record starts the fix. https://tools.openkrill.app/mailfix https://tools.openkrill.app/blog/ai-crawlers-cdn-bot-settings-b2b-software-sites https://tools.openkrill.app/blog/ai-crawlers-cdn-bot-settings-b2b-software-sites Sun, 11 Oct 2026 00:00:00 GMT We requested 10 B2B software home pages as a browser, GPTBot, ClaudeBot and PerplexityBot: no CDN refused our test requests, 1 host cut off ClaudeBot while robots.txt allowed it, and a test site from Cloudflare's docs refused all 3. On 2026-10-11, 10 of 10 B2B software home pages answered GPTBot and PerplexityBot with 200 and 9 of 10 answered ClaudeBot; robots.txt allowed all 4 bots on every site, and the 1 refusal points to a rule at the hosting company. On 2026-10-11, 10 of 10 B2B software home pages answered GPTBot and PerplexityBot with 200 and 9 of 10 answered ClaudeBot; robots.txt allowed all 4 bots on every site, and the 1 refusal points to a rule at the hosting company. https://tools.openkrill.app/aiready https://tools.openkrill.app/blog/ai-crawlers-cleaning-company-sites https://tools.openkrill.app/blog/ai-crawlers-cleaning-company-sites Sun, 11 Oct 2026 00:00:00 GMT We ran 18 cleaning-company websites through Site Check for AI crawler access and page tags: 9 answered, 1 blocks 12 of 14 AI crawlers in robots.txt, and 7 of 9 have no share image. On 2026-10-10, 9 of 18 cleaning-company sites answered our server; 1 of those 9 blocks 12 of 14 named AI crawlers in robots.txt, and 7 of 9 have no og:image. On 2026-10-10, 9 of 18 cleaning-company sites answered our server; 1 of those 9 blocks 12 of 14 named AI crawlers in robots.txt, and 7 of 9 have no og:image. https://tools.openkrill.app/aiready https://tools.openkrill.app/blog/creator-ad-program-math https://tools.openkrill.app/blog/creator-ad-program-math Sat, 10 Oct 2026 00:00:00 GMT We counted different captions and near-copy families in the top live Meta ads of Wispr Flow, Otter.ai, Granola and Superwhisper, with a reworded-copy test for the line and a Python snippet to count your own. On 2026-10-10, Wispr Flow's 59 highest-impression active Meta ads used 56 different captions; Otter.ai's 59 used 15, and one Otter.ai caption ran word for word on 29 of them. On 2026-10-10, Wispr Flow's 59 highest-impression active Meta ads used 56 different captions; Otter.ai's 59 used 15, and one Otter.ai caption ran word for word on 29 of them. https://tools.openkrill.app/blog/fframes-vs-remotion-render-times https://tools.openkrill.app/blog/fframes-vs-remotion-render-times Sat, 10 Oct 2026 00:00:00 GMT One 1080p scene built in both fframes and Remotion, checked frame by frame and timed on one Mac: 26 paired renders at 6 and 60 seconds, and 5 edit-to-video cycles per tool. On one simple 1080p clip, fframes beat Remotion in all 26 paired renders, by 1.8 to 8.1 times on the median; after a one-line edit, Remotion had a new clip ready in 9.1 seconds against 16.5. On one simple 1080p clip, fframes beat Remotion in all 26 paired renders, by 1.8 to 8.1 times on the median; after a one-line edit, Remotion had a new clip ready in 9.1 seconds against 16.5. https://tools.openkrill.app/live-card https://tools.openkrill.app/blog/cold-email-domain-check-before-after-dmarc https://tools.openkrill.app/blog/cold-email-domain-check-before-after-dmarc Thu, 08 Oct 2026 00:00:00 GMT We scored our own domain, openkrill.app, with Domain Mail Check before and after adding a DMARC record: the rules it tests, what moved the score, and what one record cannot fix. A real before and after on our own domain: 80 (almost) with no DMARC record, 96 (ready) the day we added one TXT record, and the rules that still matter. A real before and after on our own domain: 80 (almost) with no DMARC record, 96 (ready) the day we added one TXT record, and the rules that still matter. https://tools.openkrill.app/integrations/mailcheck/claude https://tools.openkrill.app/blog/greenhouse-open-jobs-and-job-age https://tools.openkrill.app/blog/greenhouse-open-jobs-and-job-age Thu, 08 Oct 2026 00:00:00 GMT How to read open jobs straight from a company's Greenhouse or Lever board, and how many of 25 sampled postings at Stripe, Airbnb, Figma, Databricks and Anthropic are 60 days old or more. Measured posting ages from the public Greenhouse boards of 5 companies, and why an old posting is a question to ask, not proof of a ghost job. Measured posting ages from the public Greenhouse boards of 5 companies, and why an old posting is a question to ask, not proof of a ghost job. https://tools.openkrill.app/integrations/company-jobs/claude https://tools.openkrill.app/blog/shopify-products-json-price-tracker https://tools.openkrill.app/blog/shopify-products-json-price-tracker Thu, 08 Oct 2026 00:00:00 GMT How to tell if a site runs Shopify, what its public products.json feed shows, and the day-over-day product, price and stock changes our Store Price Tracker measured on 3 real stores. Measured day-over-day changes on 3 real Shopify stores from their public products.json feed, plus a 1-request test for whether a site is Shopify. Measured day-over-day changes on 3 real Shopify stores from their public products.json feed, plus a 1-request test for whether a site is Shopify. https://tools.openkrill.app/integrations/store-prices/claude https://tools.openkrill.app/blog/backend-readiness-checklist https://tools.openkrill.app/blog/backend-readiness-checklist Thu, 08 Oct 2026 00:00:00 GMT The 16 backend items an AI-built app often skips, why each one matters, the primary source behind it, and which 5 Site Check reads from outside versus the 12 that stay on your own list. Which of the 16 items an outside request can really see (5, one of them only in part) and which 12 stay on your own list. Which of the 16 items an outside request can really see (5, one of them only in part) and which 12 stay on your own list. https://tools.openkrill.app/backend-checklist https://tools.openkrill.app/blog/lean-mcp-tool-answers https://tools.openkrill.app/blog/lean-mcp-tool-answers Thu, 08 Oct 2026 00:00:00 GMT What happened when we stopped echoing JSON in the text block of 113 MCP tools: the measured drop, which clients gain and which do not, and the one first-try fix that mattered most. Measured on 113 of our own tools: text fell 74%, and Claude Code users saw no change because it reads structuredContent. Measured on 113 of our own tools: text fell 74%, and Claude Code users saw no change because it reads structuredContent. https://tools.openkrill.app/integrations https://tools.openkrill.app/blog/site-check-on-our-own-site https://tools.openkrill.app/blog/site-check-on-our-own-site Thu, 08 Oct 2026 00:00:00 GMT Site Check read tools.openkrill.app from outside, found that plain http served the site, and turned clean after a one-rule fix. The run, the numbers and what it could not see. A real exposed-to-clean run on our own host: 34 requests per check, 1 medium finding, fixed with one redirect rule. A real exposed-to-clean run on our own host: 34 requests per check, 1 medium finding, fixed with one redirect rule. https://tools.openkrill.app/site-check