Real website audit example
This is an OpenForBots audit of InvoiceCraftly (invoicecraftly.com), run live against the public website on 2 October 2026 at 19:51 UTC. InvoiceCraftly then deployed fixes, and a fresh audit at 2 October 2026 at 20:44 UTC verified 3 of the findings as resolved. Every step below comes from those two recorded audits.
Decision first
The first audit found crawler access, server-delivered HTML and structured data in good shape. It flagged that the visible title led with the product category rather than the InvoiceCraftly name used in structured data, that the opening copy did not name an audience, and that llms.txt repeated two links. The re-check after the fix shows every capability as strong.
robots.txt returned 117 bytes with a single User-agent: * group allowing /. All 12 crawler tokens in the OpenForBots registry resolve to allowed at the root path (12 of 12); none has a dedicated group.
Source: https://invoicecraftly.com/robots.txt · HTTP 200About 1,470 words of server-delivered text, a title, meta description, canonical URL, lang="en" and a single H1: “Free invoice generator with logo and payment QR”.
Source: https://invoicecraftly.com/ · HTTP 2006 entities (Organization, WebSite, WebApplication, WebPage, ImageObject, FAQPage) with stable @id values; 4 use types the focused validator checks, and none is missing a core property. No malformed blocks.
Source: Homepage JSON-LD · 2 valid blocksStructured data names the organization “InvoiceCraftly”, while the title and H1 lead with the product category. The identity rule reports this as a name-consistency warning, and the wording heuristic also found identity signal is not explicit and audience language was not explicit.
Source: Homepage title, H1, opening copy and JSON-LD12 sections and 76 links (73 same-site). /privacy appears 2 times; /security appears 2 times.
Source: https://invoicecraftly.com/llms.txt · 76 linksAbout, contact, privacy and legal links are present, and the Organization declares 2 official profiles (linkedin.com, youtube.com).
Source: Homepage links and JSON-LD sameAsFrom observation to verification
Each finding follows the same seven steps. The re-check result comes from the OpenForBots comparison of the two recorded audits. Steps that have not happened are marked as such instead of being filled in for the sake of the story.
Structured and visible identity signals contain a conflict or malformed value.
Homepage identity and structured-data signals, live retrieval on 2026-10-02
This does not prove external entity recognition or profile ownership.
Align visible names, structured names, stable identifiers and official profile URLs.
Merged at 20:40 UTC. The live title changed from “Free Invoice Generator — No Signup, Logo and Payment QR” to “Free Invoice Generator — No Signup, Logo & QR | InvoiceCraftly”.
Verified resolved in a new live audit at 20:44 UTC; the website evidence changed.
Not observed since the change. The one AI-answer observation so far was recorded before it and is shown separately below.
Identity signal is not explicit; Audience language was not explicit
Server-delivered homepage wording, live retrieval on 2026-10-02
These are rule-based wording signals and do not measure provider understanding.
Make the product or organization, intended audience and practical outcome explicit in prominent server-delivered copy.
Merged at 20:40 UTC. The brand now appears in the title, and the opening copy names its audience: “freelancers and small businesses”.
Verified resolved in a new live audit at 20:44 UTC; the website evidence changed.
Not observed since the change. The one AI-answer observation so far was recorded before it and is shown separately below.
The optional file contains syntax, link or navigation opportunities.
Root llms.txt parsed findings, live retrieval on 2026-10-02
These findings do not imply provider support or visibility impact.
Correct the observed llms.txt syntax or link issues and keep the optional file concise and canonical.
Merged at 20:40 UTC. llms.txt went from 76 to 74 links, with 0 repeated destinations.
Verified resolved in a new live audit at 20:44 UTC; the website evidence changed.
Not observed since the change. The one AI-answer observation so far was recorded before it and is shown separately below.
12 registered crawler tokens were evaluated against the root path and no search-relevant crawler was found blocked.
Root robots.txt resource, live retrieval on 2026-10-02
No current search-relevant crawler restriction requiring action was identified for the tested root path. An allowed or unrestricted robots result does not prove provider crawling, indexing, retrieval, citation, ranking, recommendation or model use.
No action is required by this rule.
Nothing deployed. Broad crawler access remains a valid policy choice.
Unchanged in a new live audit at 20:44 UTC.
Not observed since the change. The one AI-answer observation so far was recorded before it and is shown separately below.
Earlier evidence
This page first showed a review of InvoiceCraftly’s repository files. The live audit replaces it as the current evidence.
External outcome
A website audit shows what a site publishes, not what an answer engine does with it. This is a separate, manually recorded observation. The answer text is not reproduced; the outcome, a summary and an accuracy check against the live page are.
The answer listed four tools. It recommended InvoiceCraftly’s E-invoice XML Viewer for Finvoice files and for reading Peppol and Finvoice in one place, and pointed to a separate validator for strict Peppol/EN 16931 compliance checks.
Seen another source cited instead of yours? AI citation forensics compares that source’s public evidence with your site — without claiming why it was chosen.
Compare a cited source with your siteInspect the same signals directly
Now check your own site
No account or email is required. Completed anonymous audit snapshots are retained for up to 30 days and then automatically deleted under the published privacy policy.