blog-post: let a post say who wrote it, in a form a machine can read

[ralyodio]

Why

A post already carries its byline in prose — "by Anthony “chovy” Ettinger" — which is exactly the form nothing can act on. Directories, feed readers and the fediverse all look for rel="me", and the template published none.

Concretely: rssamplifier indexed this blog, got the author's name from the feed's managingEditor tag, and produced an author page with an email and a GitHub link and nothing else — because the pages genuinely declared nothing else. The only outbound links on the entire blog were two to kagi.com.

What

One <p> in the post footer, inside <article>:

<p>Find me: <a rel="me" href="https://defcon.social/@chovy">Mastodon</a> ·
<a rel="me" href="https://github.com/ralyodio">GitHub</a> ·
<a rel="me" href="mailto:anthony@profullstack.com">email</a></p>

rel="me" outranks h-card, JSON-LD sameAs and ordinary footer links in every identity extractor I know of, because it is the one a publisher has to state deliberately.

Verification

Run against rssamplifier's production identityFromHtml, on the live page with the same block already applied:

"profiles": [
 { "network": "fediverse", "url": "https://defcon.social/@chovy", "source": "rel-me" },
 { "network": "github",    "url": "https://github.com/ralyodio",  "source": "rel-me" },
 { "network": "email",     "url": "mailto:anthony@profullstack.com", "source": "rel-me" }
]

All three land as rel-me. Both handles were confirmed to be the person rather than the company: defcon.social/@chovy titles as "Anthony Ettinger (@chovy@defcon.social)", github.com/ralyodio as "ralyodio (Anthony Ettinger)".

Personal accounts only. The company X and YouTube accounts are deliberately absent — filing an org account under a person is worse than finding nothing, and good extractors reject them anyway.

Kept smolweb-valid: one paragraph, closed, no new elements.

pnpm test 132 passing · pnpm typecheck clean.

Note

This changes the template for new posts. The live blog's index.html — which is what a crawler actually fetches for the feed's site — was updated separately and directly, since that directory is not this repo.

🤖 Generated with Claude Code

[github-actions]

ThreatCrush Security Scan

2 finding(s)

LOW: 2

Severity	Rule	Location
LOW	insecure-temp-file	test/blog.test.ts:56
LOW	insecure-temp-file	test/blog.test.ts:57

Snippets are redacted; ThreatCrush never prints matched credential material.