Skip to content
SkillModel-led1,377 tokens in use

seo-flow

Work through the FLOW loop of Find, Leverage, Optimize and Win with a library of 41 evidence-led SEO prompts that the skill picks and applies by context.

31,135 stars on the repository MIT LicenseAgriciDaniel/claude-seo · skills/seo-flow
/plugin marketplace add AgriciDaniel/claude-seo /plugin install claude-seo@agricidaniel-claude-seo
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing seo-flow costs

80tokens idle
1,377tokens in use
6,476tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

shelldetectednetworkdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter15/15
  • Structure17/25
  • Examples9/15
  • Output10/15
  • Size15/15
  • Resources15/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • File that was not read−15

    a file the skill ships that was not read, so what it does is unknown

    references/flow-prompts.lock

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community72
  • Maturity32
  • Momentum40

seo-flow brings the FLOW framework into Claude Code. FLOW is an evidence-led SEO operating model for the AI-search era, and the skill ships its prompt library: five find prompts, one leverage prompt, 21 optimize prompts, three win prompts and eleven local prompts.

Each stage command loads only the prompts it needs and applies them to a URL or topic. The optimize stage picks two or three prompts by industry, prior analysis and URL type, and every run prints the CC BY 4.0 attribution before the analysis.

What You Get

  • A stage overview and a one-line description of each FLOW stage
  • Applied find prompts for keyword research, gap analysis and SERP intent mapping
  • Two or three context-matched optimize prompts, with the reason for each choice
  • Win-stage prompts for bottom-of-funnel content, conversion and a dual-surface scorecard
  • A prompt index and a sync summary of files added, updated and unchanged

Common Use Cases

  • Run structured keyword discovery for a new topic
  • Choose the right optimization prompts after a technical or content audit
  • Review local SEO: Google Business Profile, titles and local audits
  • Keep a team’s SEO prompts current with the FLOW repository

What Makes It Unique

It is a prompt library with no scoring of its own. Selection is limited to two or three optimize prompts so output stays focused, each run points to the related claude-seo skill for deeper data, and the attribution notice cannot be dropped.

Under The Hood

The skill reads reference files and prompt files from its own folder on demand and applies them to a URL or topic. Only the sync command runs a program: it starts a Python script through the claude-seo launcher in a shell, which pulls the latest prompt files from the FLOW repository on GitHub over the network and writes them into the skill’s references folder. Network errors and authentication problems are reported with a pointer to the GitHub CLI. The folder also ships a lock file that the server could not read, so what it holds is unknown. Fetched content is treated as untrusted data. The frontmatter does not declare the shell, network or file access.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers