seo-content-brief — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited seo-content-brief (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You generate a content brief that ranks AND gets cited by AI. Modern briefs are no longer "keyword + word count + headings." They are entity-aware, intent-mapped, E-E-A-T-aware, and AI-citation-targeted.
Why this differs from a 2023-era brief:
============================================================ === PRE-FLIGHT === ============================================================
Verify:
Recovery:
============================================================ === PHASE 1: SERP + AI CITATION ANALYSIS === ============================================================
For the target query, gather signals from THREE surfaces:
Output brief/01_competitive_landscape.md with side-by-side citation tables.
VALIDATION: Per-engine citation list has ≥ 3 URLs. Overlap with organic top-10 noted.
FALLBACK: Without Perplexity/ChatGPT API access, capture manually via the user's browser (provide them a copy-paste prompt). Less precise but unblocks the brief.
============================================================ === PHASE 2: ENTITY EXTRACTION === ============================================================
From the top 10 organic + AI-cited sources, extract every recognized entity (proper noun mapped to a Knowledge Graph node).
Bucket entities by type:
Output brief/02_entities.md ranked by co-occurrence frequency across sources.
Identify:
VALIDATION: Entity list has ≥ 25 candidates. Must-mention list has ≥ 10 entities.
============================================================ === PHASE 3: INTENT-MAPPED HEADING STRUCTURE === ============================================================
Generate the H2/H3 outline. Each heading is a question the reader (or AI engine) is trying to answer.
Default outline patterns by intent:
Informational ("what is X", "how does X work"):
Commercial-investigation ("best X", "X vs Y", "X review"):
Transactional ("buy X", "X pricing", "X near me"):
Output brief/03_outline.md.
VALIDATION: Outline is intent-matched. Each H2 maps to a specific user question or buyer-journey stage.
============================================================ === PHASE 4: FAQ MINING === ============================================================
Compile the FAQ section from multiple sources:
Dedupe, rank by frequency, output as a 5–10 question FAQ section with answer drafts (1–3 sentences each).
VALIDATION: FAQ has ≥ 5 questions, each with a draft answer ≤ 60 words.
============================================================ === PHASE 5: INTERNAL LINKING TARGETS === ============================================================
From the user's existing top pages, identify:
Format each as:
- [{Anchor text}](https://example.com/path) — Why: {one-sentence rationale}VALIDATION: Each suggested link has a specific anchor text and rationale, not just a URL.
FALLBACK: Without site access, suggest internal-link templates ("link to your pillar page on {X} topic") and let the writer fill specifics.
============================================================ === PHASE 6: E-E-A-T REQUIREMENTS === ============================================================
Specify what makes this content citable, not just rankable:
/author/{slug}, Person schema, ≥ 3 sameAs.dateModified schema and visible "Updated YYYY-MM-DD" in DOM.Mark each as REQUIRED or RECOMMENDED based on competitiveness of the query (high-competition → all REQUIRED).
VALIDATION: List is specific to this brief's topic, not generic boilerplate.
============================================================ === PHASE 7: THE FINAL BRIEF === ============================================================
Combine all phases into brief/{slug}.md:
# Content Brief — {Target Query}
**Generated:** {date}
**Intent:** {informational / commercial-investigation / transactional}
**Target length:** {word count based on competitor median ±10%}
**Author requirement:** {Person entity profile complete: yes/no — must be yes to publish}
## TL;DR for the writer
- Lead with the direct answer in the first 60 words.
- Cover these MUST-MENTION entities: {list of 10}
- Hit ≥ 15 total recognized entities (see entities.md for the full pool).
- Internal-link to: {3 pages}
- FAQ section is required (5+ questions from FAQ.md).
- Author must be {Name}; if not, set up the Person entity first.
## Outline
{from Phase 3}
## Entities to include
{from Phase 2, must-mention bucket}
## FAQ section
{from Phase 4}
## Internal Linking
{from Phase 5}
## E-E-A-T requirements
{from Phase 6}
## Competitor benchmarks
- Median word count: {N}
- Median entity count: {N}
- Median schema types: {Article, FAQPage, ...}
## Schema to ship with the page
- Article (required)
- FAQPage (if FAQ section is present)
- BreadcrumbList
- {anything else specific to this topic — HowTo, Product, etc.}VALIDATION: Brief is ≤ 6 pages of markdown. A competent writer should be able to produce the article from the brief alone without further research.
FALLBACK: If brief is too thin (insufficient SERP signal — emerging topic), say so explicitly and recommend supplementing with subject-matter-expert interview.
============================================================ === SELF-REVIEW === ============================================================
Score 1–5:
Common gap: brief loses entity discipline at writing time. Cap the brief by reminding the writer in the TL;DR to count entities before publishing.
============================================================ === LEARNINGS CAPTURE === ============================================================
Append to ~/.claude/skills/seo-content-brief/LEARNINGS.md:
============================================================ === STRICT RULES === ============================================================
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.