InspiMarkBot
InspiMark helps people save and organize links. Our server retrieves titles, previews, and readable article text when people paste, share, or refresh links. It also retrieves full text for subscribed RSS feeds and for Discover saves. Feed enrichment can run in the background before an individual article is saved.
Recognizing our requests
Our default HTTP user agent is InspiMarkBot/1.0 (+https://inspimark.com/bot).
Our article fetcher can authenticate requests using Web Bot Auth. When signing is enabled, Signature-Agent identifies https://inspimark.com. The public key directory is at /.well-known/http-message-signatures-directory; it is available once a signing key is configured.
Publisher controls
The generic article fetcher checks robots.txt before requesting an article, including at redirect destinations. For user saves and refreshes, it follows rules naming InspiMarkBot; wildcard rules and Crawl-delay do not apply to those requests. For background RSS enrichment and Discover, it follows named rules, falls back to * when no named group exists, and applies Crawl-delay up to sixty seconds. An applicable Disallow rule prevents both article and preview retrieval.
User-agent: InspiMarkBot Disallow: /
Ordinary 4xx responses and HTML catchall pages at robots.txt permit an article request. A robots challenge also permits a user save; background fetching defers when a successful robots response contains a challenge. Server errors and network failures defer both kinds of requests. A 429 response defers requests using Retry-After.
Each server process spaces requests to an origin at least one second apart. Waiting user requests take priority over background work without adding request capacity. Robots rules are cached for up to one hour. Retry-After delays are limited to ten minutes; missing or malformed values use thirty seconds. Capture requests have a deadline, so longer waits require a later attempt.
Optional Firecrawl recovery is off by default. When enabled on selected public article hosts, a failed user save or refresh can use it to retrieve readable content, subject to daily call limits. We check our named publisher rules before that request and require the actual final destination to stay on the approved host and pass those rules before accepting the article. Firecrawl operates its own fetching infrastructure; those requests are not authenticated with our Web Bot Auth key, and validating the returned destination does not establish which destinations its internal redirects contacted. We do not send browser cookies, passwords, or account sessions. Background RSS and Discover fetching do not use this recovery path.
When automatic capture still fails or requires login, the user can open the page themselves and choose to save the displayed article with the InspiMark browser clipper. Specialized media integrations and RSS feed XML retrieval use separate fetch paths.
Contact
For questions about requests or to report a problem, contact support@inspimark.com. Please include the affected domain, approximate request time, and relevant request headers. Do not include passwords or session cookies.