pcamarajr

Create content of any configured type from a topic or from the content backlog. Orchestrates research, writing, style review, glossary creation, linking, and reindexing as focused sub-agents. Hooks handle tracker updates and build verification automatically.

content-seo setup wizard. Run /init to see status, or /init [round] to run a setup step. Rounds: project, credentials, strategy.

Site-level SEO intelligence. Routes to one of five workflows — status (site-wide GSC performance), opportunities (pages ranking 4–20), diagnose (deep-dive a page with GSC data + fixes), brief (keyword research before writing), or coverage (audit GSC indexing errors from a CSV export).

content-ops setup wizard. Run /init to see status, or /init [round] to run a setup step. Rounds: project, content-types, style, strategy, infra, images, link-building.

Identify content gaps and add suggestions to the backlog. Accepts an optional freeform payload to narrow scope. Without a payload, surveys the full corpus against pillars. Powered by the backlog-suggester agent.

Loads and applies the project's configured style guide. Hard stops if config or any required piece is missing. All style rules come from the guidelines file set during /init style.

Review content for tone, style, structure, glossary coverage, and linking completeness against the style guide.

Translate content to a target language using localization (not direct translation). Handles batch translation with glossary, linking, and tracker updates. Reads paths from plugin config.

Help a QA team use Scout — browser QA where natural-language scenarios are AI-verified once and then replayed deterministically. TRIGGER when the user mentions Scout, a `.scout.md` scenario, `scout.config.json`, `scout go` / `scout init` / `scout report`, browser/E2E/QA testing of a user flow, or asks to author, verify, replay, or triage a browser test scenario. Do NOT trigger for unrelated unit tests or non-browser work.

Autonomously check the health of the personal-knowledge memory system AND validate whether the autolearn improvements are actually working (not just active). Runs a blind LLM-as-judge efficacy eval against a frozen probe set, scores re-rank vs raw retrieval, tracks the trend in a ledger, and judges how well the dashboard surfaces that progress. Invoke when the user wants a memory-system health + efficacy report, asks "is autolearn working / improving", or on a schedule (cron-ready).