Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This collection provides SEO specialists and content strategists with tools to maintain site health and refine editorial planning. It automates the detection of keyword cannibalization and performs full-site audits to ensure content remains distinct and effective.
The suite extends into the production phase, offering capabilities for generating detailed content briefs, managing editorial calendars, and converting written posts into narrated audio formats.
WHO IT'S FOR
SEO Specialists
detect keyword overlap and site health issues
Content Strategists
plan editorial calendars and content briefs
Digital Marketers
maintain durable brand voice and editorial context
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Use the blog-outline skill to create a lightweight SERP-informed outline for [TOPIC]. Use [TARGET KEYWORD] if it differs from the topic and [INFORMATIONAL, COMMERCIAL, OR TRANSACTIONAL] as the search intent; infer omitted values from context. Include title suggestions, target parameters, an H2/H3 skeleton with section estimates, optional FAQ slots, internal-linking zones, and 3–5 content gaps. Do not write the full article. Save the outline to [OUTPUT PATH], or use outlines/[slug]-outline.md if no output path is supplied.
Paths to 5–10 representative posts from the same author, brand, or editorial voice
Use the blog-style skill to analyze 5–10 representative [MARKDOWN, MDX, OR TEXT FILE PATHS, OR A DIRECTORY] from the same author or editorial voice. Produce a concise voice profile covering the measured style signals and include a VOICE.md-ready block in the response. Do not overwrite or create files.
Previous-period GSC JSON page export from an adjacent period of similar length
Confirmation of matching dimensions and filters, if available
Use the blog-decay skill's default offline workflow to compare [CURRENT GSC JSON EXPORT] with [PREVIOUS GSC JSON EXPORT]. Use clicks and the default 20% decline threshold. Report each flagged page and its severity. Treat pages missing from the current export as needs_validation unless identical filters, row limits, dimensions, and URL status have been confirmed. Do not recommend refresh, query-shift, consolidation, redirect, or pruning actions yet; instead, list the additional checks needed for action selection, including indexation, canonical status, query loss, internal links, backlinks, seasonality, and business value. Do not modify content.
Claude
Gemini CLI