Standard.site Audit

AT Protocol (atproto) long-form publishing requires precise record management to ensure posts appear in search and reader apps. When blog posts fail to render in Bluesky link cards or indexers, the cause is usually missing textContent or broken verification tags. This skill provides a suite of Python tools to audit and repair standard.site (site.standard) lexicon implementations.

What it does

• Audit implementations by validating .well-known discovery files and bidirectional document link tags.
• Identify critical errors including missing required fields, dead links, and empty text content records.
• Remediate PDS records by backfilling plaintext bodies, setting discovery flags, and updating branding.
• Prune dead records that point to 404 pages to maintain firehose health and indexer accuracy.
• Generate reports in Markdown or HTML formats to prioritize fixes across single or multiple domains.

How it works

1. Run audit.py with --check-links to scan the domain and generate a structured JSON findings file.
2. Execute report.py to transform findings into a prioritized list of remediation steps.
3. Use remediate.py with the textcontent command and an app password to backfill missing post data.
4. Set the discover flag and update branding fields to ensure the publication is visible in discovery feeds.

Frameworks & tools

Python 3.8+ (stdlib only). No external package dependencies. Supports AT Protocol, Bluesky API, and PLC directory.

Why this beats prompting it yourself

Manually managing atproto records requires strict adherence to CBOR size limits and complex batched applyWrites logic. This tool handles record merging, UTF-8 byte budgeting, and PDS rate limiting out of the box.

Use cases

• Fix Bluesky link cards that are missing descriptions or reading time estimates.
• Synchronize a Hugo or Jekyll blog with a PDS after changing URL structures.
• Audit a network of publications to ensure discovery settings are correctly enabled.
• Repair an existing archive where records were created without textContent payloads.

Known limitations

Requires an AT Protocol app password for write operations. Scraping for textContent is a fallback that may pick up page navigation if the site lacks proper article tags.