메뉴
How to present the deterministic AI-SEO audit returned by audit_page. Use when reporting a page's readability and citation-readiness for AI answer engines (ChatGPT, Claude, Perplexity).
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
How to present the deterministic AI-SEO audit returned by audit_page. Use when reporting a page's readability and citation-readiness for AI answer engines (ChatGPT, Claude, Perplexity).
Migrate Mastra code-based agents (agents built with new Agent(...) and registered in a Mastra({ agents }) map) to the file-based convention: one directory per agent under src/mastra/agents. Use when the user wants to convert or refactor Mastra agents to the per-directory convention, mentions Mastra "file-based agents" or agentConfig, or wants to split a big src/mastra/index.ts or agents.ts into one folder per agent — even if they do not name this skill. Requires @mastra/core 1.48+.
Use when building, debugging, reviewing, or documenting Flue agents, workflows, channels, skills, tools, sandboxes, targets, routing, persistence, observability, or CLI usage; routes coding agents to version-matched Flue documentation through the CLI.
How to safely operate the sandboxed workspace. Use for any multi-step task involving files or shell commands.
Iterative, self-evaluating web research. Use when answering open-ended questions that require gathering and verifying information from multiple sources.
| name | seo-audit |
| description | How to present the deterministic AI-SEO audit returned by audit_page. Use when reporting a page's readability and citation-readiness for AI answer engines (ChatGPT, Claude, Perplexity). |
How well a page can be crawled, chunked, understood, and cited by AI answer
engines — not classic blue-link SEO. The audit_page tool computes this
deterministically: page data flows through context.dev, then a fixed rubric of
~30 checks scores six categories into a 0–100 total. Your job is to present
that result, not to re-score it.
audit_page returns JSON with:
score (0–100) and band ({ label, interpretation }).stats — words, headings, links, externalLinks, jsonLdBlocks, char counts.categories[] — each { id, name, score, maxScore, naExcluded, items[] }.categories[].items[] — each check: { id, label, status, evidence, recommendation, score, maxScore }. status is pass | partial | fail | na.topPriorities[] — the highest-impact failing/partial checks, pre-sorted.agentPrompts — { full, topIssues[] }, ready-to-run fix prompts.diagnostics — context.dev fetch status and JSON-LD parse failures.| ID | Category | Max |
|---|---|---|
| A | Technical AI Crawlability | 28.4 |
| B | Content Structure & Chunking | 25.4 |
| C | Structured Data / Schema | 13.4 |
| D | E-E-A-T & Entity Authority | 21.4 |
| E | Off-site / Citation Surface Presence | 8.4 |
| F | Measurement & Governance | 3 |
Scoring (already applied by the tool): pass = full weight, partial = half,
fail = 0, na = excluded from the denominator. Bands: Critical ≤40 ·
Below average ≤55 · Average ≤70 · Strong ≤85 · Excellent >85.
Report, in order:
score/100, its band.label, and the band.interpretation.score/maxScore and the
pass/partial/fail counts.topPriorities
and the matching items (cite each check's id, evidence, and
recommendation).agentPrompts.full verbatim in a
copy-paste block.Report only the checks and numbers the tool returned. Never invent or recompute scores, checks, evidence, or content.