Conciseness

The body is dense with non-obvious domain specifics (E-E-A-T dimension definitions, per-industry threshold table, citation-ledger CLI usage, anti-pattern list) that Claude would not already know, so most tokens earn their place. It sits at the 4 anchor rather than 5 due to minor trim candidates: the 'Trigger Phrases' section duplicates the frontmatter description, the 'Both can coexist...' prose under 'Distinct From SEO' explains concepts Claude already knows (SEO rewards backlinks/keywords), and the version/source footer adds metadata tokens; it is above the 3 anchor because there is no substantial padded explanation.

Actionability

Quotes: '$ python3 scripts/aeo_audit.py --url https://example.com/blog/post', '$ python3 scripts/aeo_optimizer.py --input post.md --mode balanced --output post-aeo.md', 'python3 scripts/aeo_audit.py --url <url> --industry healthcare', and tracker add/report invocations — concrete, flag-level commands for the common cases, plus a full example output report. This matches the 4 anchor ('concrete code or commands with minor gaps') rather than 5 because the optimizer's output is never exemplified, the 'query_research mode' is mentioned only as scaffold-only with no usage, and the audit report example contains '[...]' placeholders.

Workflow Clarity

The numbered 0-4 workflow is clearly sequenced with a real command and expected output ('→ markdown report with composite score + 4-dimension breakdown') per step, and step 0 is an explicit gating pre-check ('a blocked GPTBot/PerplexityBot/ClaudeBot/Google-Extended is the first fix, always'). This matches the 4 anchor; it is not 5 because there is no explicit validation or error-recovery step (e.g., reviewing the optimizer diff before publishing, or what to do when the audit scores below an industry threshold), and it is above the 3 anchor because checkpoints are not merely implicit — the pre-flight gate and per-step expected outputs are stated.

Progressive Disclosure

References are clearly signaled with one-line descriptions ('references/aeo_eeat_canon.md — E-E-A-T methodology, industry thresholds, anti-patterns') and are one level deep, but none of the referenced files exist in the bundle — there are no references/, scripts/, or assets/ directories, so all 5 reference paths and all 3 script paths are dangling. Additionally, content that the reference descriptions say lives in those files (the industry-threshold table duplicating 'aeo_eeat_canon.md' territory, and the Anti-Patterns section overlapping the canon's 'anti-patterns') is inlined in the body. Good signaling but broken navigation and duplicated placement matches the 3 anchor ('structure present but content that should be separate is inline / references do not resolve); it is above 2 because the body itself is well sectioned and not an unstructured monolith, and below 4 because referenced paths do not resolve to real files.

Specificity

Quotes: 'optimize content to be cited by AI language models', 'auditing existing content for E-E-A-T signals', 'tracking which pages get cited by which LLMs', 'Output is a markdown audit report (default) or JSON for pipeline integration'. Several concrete, distinct actions are named with a concrete output format, matching the 'several specific actions; minor gaps' anchor; it falls short of the 5 anchor because 'building a citation-friendly content strategy' and 'optimize content' remain moderately abstract, and it is above the 3 anchor because more than 1-2 actions are covered.

Completeness

Quotes: what — 'optimize content to be cited by AI language models... as authoritative sources... audits, optimizes, and tracks' (implied by the action list) plus explicit output format; when — 'Use when planning content for AI-first search audiences, auditing existing content for E-E-A-T signals, tracking which pages get cited by which LLMs, or building a citation-friendly content strategy' followed by explicit trigger phrases. Both what and when are explicitly and concretely answered, matching the 5 anchor exactly; the 4 anchor's 'when could be more explicit' does not apply.

Trigger Term Quality

Quotes: 'AEO audit', 'optimize for ChatGPT', 'get cited by Perplexity', 'LLM citation strategy', 'answer engine optimization', 'content for AI search', 'E-E-A-T audit' — plus when-clauses like 'tracking which pages get cited by which LLMs'. Good, natural keyword coverage with several phrasings users would actually say; not the 5 anchor because common synonyms such as 'generative engine optimization' / 'GEO' and 'AI SEO' are missing.

Distinctiveness Conflict Risk

Quotes: 'Distinct from SEO — AEO optimizes for citation in LLM-generated responses, not search rankings'. The description explicitly draws the boundary against the closest adjacent skill (SEO) and uses niche-specific triggers ('AEO audit', 'get cited by Perplexity', 'E-E-A-T audit') that no generic skill would claim, matching the 'clear niche with distinct triggers; minimal conflict risk' anchor rather than the 4 anchor, which requires residual overlap with closely related skills.