AI Agent Workflow Collection

Generate an "alternatives to [Competitor]" page that captures searchers actively shopping away from a named competitor. Listicle-style — ranks the site alongside 4–7 other named alternatives, with the site positioned as best-fit for a specific use case (not always 1). Picks the keyword from service-keywords.csv (rows with intent=alternative) or user input, runs SERP analysis, satisfies Tier 1 on-page SEO, validates schema, and ships per the site's publishing method. Use when the user types /alternative or asks for an "alternatives to" page.

Monthly cron — finds pages on the open web that link to defunct competitor URLs where the site has a live replacement page, queues outreach to pitch the site's URL as the broken-link replacement. Pulls broken-backlink data from SEMrush (or Ahrefs/Moz as fallback) for each top-5 competitor, cross-references with the site's published content, surfaces high-probability "replacement-ready" opportunities. Highest-conversion outreach type — link reclamation pitches convert ~10× higher than cold-pitches because the publisher already has a broken link to fix. Use when the user types /broken-backlinks or when the monthly cron fires.

Generate a head-to-head comparison page ("[Site] vs [Competitor]" or "[Product A] vs [Product B]") for a chosen managed site. Captures decision-stage searchers comparing the site against a named alternative. Picks the comparison from service-keywords.csv (rows with intent=comparison) or user input, runs SERP analysis on top 3 comparison pages, builds a feature-by-feature comparison table, balances honesty (real limitations) with conversion (clear differentiators), satisfies Tier 1 on-page SEO, and ships per the site's publishing method. Use when the user types /comparison or asks for a "vs" page.

Monthly cron — reverse-engineers the backlink graphs of the site's top-5 ranking competitors to surface high-probability link-acquisition targets. Pulls each competitor's top referring domains via SEMrush (or Ahrefs/Moz fallback), filters to domains NOT yet linking to the site, prioritizes domains linking to multiple competitors (= proven willingness to link to this vertical), scores by DA + topical relevance + outreach probability, and queues opportunities for user review in inbox/. Use when the user types /competitor-backlinks or when the monthly cron fires.

Pull today's journalist-query digest from the site's configured source (Featured.com, Source of Sources, Qwoted, Help a B2B Writer, or HARO if revived), match queries to the site's documented expertise (per site-info.md + stats.md + stories.md), auto-draft pitch responses for high-match queries, and queue them for user approval in inbox/. Runs daily as a cron. Highest-velocity backlink-acquisition channel — journalist queries move on 4–24 hour windows, so daily cadence is mandatory. Use when the user types /haro or when the daily cron fires.

Use when the user wants to see which MCP connectors are active, which are available to add, and which skills each connector unlocks. Provides a full integration status dashboard.

Drive a Lovable round-trip via Chrome DevTools MCP. Reads a prepared prompt from the site's drafts/ folder, navigates the user's local Chrome to the right Lovable project, pastes the prompt atomically (DataTransfer-based, bypassing Cmd+V UTF-8 mangling), waits for Lovable to finish building, captures the diff, runs heuristic verification, and either queues for human approval (default mode) or clicks Publish → Update directly (--auto-approve, subject to eligibility rules). After publish, runs curl-based verification on the live URL. Interactive-only — requires a local Claude session with Chrome MCP connected. Use when the user types /lovable-deploy or asks to ship a queued Lovable deploy.

Generate N service pages from a matrix CSV (e.g., services × cities, services × industries, services × company-size). Each output cell is a unique service page passing all six anti-doorway-page rules from SEOGUIDE Section 4.1. Refuses to ship any cell that can't satisfy uniqueness. Enforces footprint volume caps from Section 4.2 (service-area: 30 max initial; multi-location: 50 cap). Produces drafts to drafts/programmatic-{matrix-name}/ with one draft per cell. Use when the user types /programmatic-batch or asks to scale service-page production via templates.

Refresh a site's SEMrush snapshot — domain authority, organic traffic, keyword count, top keywords, top competitors, backlink snapshot — and append the result as a dated baseline file in the site folder. Use when the user types /semrush-baseline {site} or when the quarterly cron fires.