Daily Ahrefs backlink movement digest in Slack

Every weekday at 9am ET (cron trigger), I want an agent to post a backlink movement digest into a Slack channel for one or more target domains I configure. The whole point of making this an agent rather than a raw dump is that I want it to use judgment to rank what actually matters and drop the noise.

Configurable inputs the workflow should accept:

- target_domains: one or more domains to monitor, e.g. ["mycompany.com"]. Required.

- slack_channel: the Slack channel ID or name to post the digest into. Required.

- min_domain_rating: minimum Domain Rating to consider a backlink worth surfacing. Default 20.

- max_items_per_section: how many items to include in each section of the digest. Default 5.

For each target domain, in parallel, the agent should do the following:

1) Call the Ahrefs List Backlinks operation to find newly discovered backlinks. Use mode=domain and a where filter that keeps only rows where is_spam = false, domain_rating >= min_domain_rating, and first_seen is within the last 24 hours. Order by domain_rating desc. Select at least: url_from, url_to, domain_rating, traffic, anchor, first_seen, is_dofollow, link_type. Remember Ahrefs requires the select parameter on list endpoints and the where parameter uses JSON filter syntax, not a string expression.

2) Call List Backlinks again with a filter for lost links: last_seen within the last 24 hours and the row marked as lost. Keep the same is_spam = false and domain_rating >= min_domain_rating filters. Select url_from, url_to, domain_rating, anchor, last_seen.

3) Call List Broken Backlinks to find external sites linking to URLs on the target that now return 4xx or 5xx. These are redirect candidates. Filter to is_spam = false and a sensible Domain Rating floor. Capture url_from, url_to, http_code, domain_rating.

4) Call List Referring Domains to compute today's net referring domain change versus yesterday for the one-line trend headline. Use the where filter to exclude spam.

Then synthesize a single Slack message per run (not one per domain). Group sections by domain inside that message. Use Slack mrkdwn formatting (*bold*, _italic_, <url|text>). The message structure should be:

- A bold one-line trend headline per domain, e.g. "*mycompany.com:* net +3 referring domains today, DR-weighted +12".

- *New backlinks worth a thank-you:* up to max_items_per_section highest-DR new links, each a one-liner with referring domain, DR, anchor text (italicized), and the page they point at. Phrase as outreach candidates.

- *Lost links worth reclaiming:* up to max_items_per_section highest-DR lost links with referring domain, DR, anchor text, and what they used to point at. Phrase as reclamation candidates.

- *Broken inbound links to fix:* the broken target URL, the referring page linking to it, and a suggested action (set up a 301 redirect to the closest live URL).

- A closing one-line totals summary across all domains.

Post the message using Slack Bot's Send a Message operation into the configured channel. If a domain had zero new, zero lost, and zero broken items today, omit that domain's section entirely. If every domain is quiet, post a single short line like: "No backlink movement above DR <threshold> today across <N> domains." That way the channel stays clean but the team knows the check ran.

Judgment guidance: do not just dump the API rows. Rank within each section by domain_rating, but prefer a lower-DR link if the referring site or anchor text is unusually interesting (an editorial mention from a known publication beats a generic directory listing). Keep the whole message under about 30 lines.