discover
無料Discover and normalize entities for programmatic SEO from a seed keyword or CSV file. Writes normalized entities to data/entities.json. Use when the user wants to find, import, or expand a list of SEO targets.
日本語の概要は準備中です。原文の説明を表示しています。
Enrich discovered entities with AI-generated content using Claude via the Anthropic SDK. Outputs structured JSON per entity for downstream use in page templates. Requires ANTHROPIC_API_KEY (or use --dry-run to preview prompts).
日本語の概要は準備中です。原文の説明を表示しています。
インストール方法を見るインストールする前に、エージェントに与えられる指示の中身を確認できます。
Generate AI-assisted content JSON for each entity using Claude, ready for injection into page templates.
Invoke the sophon skill first. It contains the entity model, enrichment conventions, and the Context Gathering Protocol.
Follow the protocol before doing any work:
## Sophon Project Context section..sophon.md in the project root.sophon teach now — do NOT skip this step.ANTHROPIC_API_KEY must be set in the environment (unless using --dry-run). Enrichment will be retried with exponential backoff on transient errors (429/500/502/503/529).
export ANTHROPIC_API_KEY=your_key_here
npx @sophonn/sophon enrich
# Preview prompts without calling the API
npx @sophonn/sophon enrich --dry-run
# Custom output directory
npx @sophonn/sophon enrich --enrich-output ./data/enriched
# Specify entity source
npx @sophonn/sophon enrich --entities ./data/entities.json --enrich-output ./data/enriched
One JSON file per entity written to data/enriched/ (or --enrich-output). Each file contains structured content fields that map to the // TODO sections in generated page templates. An enrichedAt timestamp is set for downstream freshness tracking.
--max-retries)--dry-run outputs the system and user prompts without calling the API, useful for reviewing what would be sentANTHROPIC_API_KEY is missing (and not dry-run), exit early with a clear message mentioning --dry-run.// TODO markers visible.--force is passed.import { enrich } from "@sophonn/sophon";
await enrich({
entities: result.entities,
output: "data/enriched",
dryRun: false, // set true to preview prompts without API calls
maxRetries: 3, // exponential backoff retry count
});
Wire the enriched JSON files into the generated page templates by reading them during build time or at request time, depending on the host framework's data fetching model.
まだレビューはありません。使ってみた感想をお寄せください。
ログインなしで書き込めます。使ってみた感想、質問、動かなかった環境などを気軽にどうぞ。
まだコメントはありません。最初のコメントを書いてみませんか。
Discover and normalize entities for programmatic SEO from a seed keyword or CSV file. Writes normalized entities to data/entities.json. Use when the user wants to find, import, or expand a list of SEO targets.
日本語の概要は準備中です。原文の説明を表示しています。
Generate one static framework page per entity using the appropriate Sophon adapter. Supports Next.js, Astro, Nuxt, SvelteKit, and Remix. Use when the user wants to scaffold programmatic SEO pages.
日本語の概要は準備中です。原文の説明を表示しています。
Pull real performance data from Google Search Console, analyze weak-performing pages, and generate actionable optimization recommendations. Optionally applies safe auto-fixes to Sophon-generated content.
日本語の概要は準備中です。原文の説明を表示しています。
Execute the full Sophon pipeline: discover → generate → technical → enrich. Use when the user wants to run all steps in one command or build a complete pSEO surface from scratch.
日本語の概要は準備中です。原文の説明を表示しています。
Master context for Sophon — an agentic programmatic SEO toolkit. Teaches the framework, entity model, adapter system, and operating rules. Invoke this skill before using discover, generate, technical, enrich, optimize, or run so you have the full context. Call with 'teach' to run the onboarding interview and create .sophon.md.
日本語の概要は準備中です。原文の説明を表示しています。
Generate technical SEO assets: sitemap.xml (with sitemap index for large sites), robots.txt, per-entity schema markup (BreadcrumbList, AggregateRating), configurable internal link graph, and hreflang scaffold. Use when the user wants to add or update sitemap, robots, structured data, internal linking, or multilingual hreflang tags.
日本語の概要は準備中です。原文の説明を表示しています。