Conciseness

The body never explains concepts Claude already knows, but the same discipline sentences are repeated many times: "every number labeled Measured / proxy / User-provided" appears in the intro, the contract's expected output, the done-when clause, step 7, and the termination rule; "never adjudicates" appears three times (contract writes, step 7, Save Results); the scope-guard and Reference Materials sections restate the same sibling-skill routing twice. This matches the 3 anchor ("mostly efficient but... could be tightened") — it is not 4 because the redundancy is systematic, not a few trimmable lines.

Actionability

Concrete, executable guidance throughout: named connector scripts with flags ("scripts/connectors/tavily.py --answer", gdelt.py, bluesky.py, pageviews.py), exact memory paths (memory/narrative-registry/canon.md, memory/events/claims.ndjson), a ready-to-use Quick Start prompt set, and a precise echo-rate definition with the declare-method-first rule. Not 5 because no actual example invocations or a sample report fragment are shown — the reader must infer argument shapes for the connectors and the report layout is only described, not exemplified.

Workflow Clarity

A clear 7-step sequence with explicit validation checkpoints: "If no current non-forked canon exists, stop with NEEDS_INPUT", the method-must-be-declared-before-computing defect rule, the trend-restart guard ("do not merge a new panel into an old trend line"), the [needs source] evidence gate routing to claims.ndjson, and a done-when checklist closing the loop. This is a read/measure skill so destructive-operation caps do not apply; it matches the 5 anchor's explicit validation and checklist criteria.

Progressive Disclosure

Well-organized sections with a Reference Materials list that signals each one-level-deep link and what it is for (tale-benchmark.md, stimulus-binding.md, CONNECTORS.md, SECURITY.md, sibling SKILL.md files). No bundle files ship with this skill (references/, scripts/, assets/ are absent), so the score rests on the body's own structure, which is good but not the 5 anchor: some detail (handoff-summary format, save-results template) is delegated to an external skill-contract.md outside this bundle, and several inlined routing/handoff statements duplicate material that belongs in those referenced files.

Specificity

Lists multiple concrete actions with the exact machinery for each: "echo rate (overlap of market language with the narrative-registry canon lexicon, method declared)", "AI-answer perception via tavily.py --answer (proxy-labeled)", "share-of-voice on a locked competitor panel (reusing share-of-voice-tracker)", "resonance signals from bluesky.py / gdelt.py / pageviews.py" — comprehensive and concrete throughout. It is dense to the point of being a single run-on sentence, but every clause names a specific capability, not padding, so the verbosity guideline does not pull it below the 5 anchor.

Completeness

Explicitly answers both questions: what ("produces a resonance report — echo rate..., AI-answer perception..., share-of-voice..., and resonance signals..., every number labeled Measured / proxy / User-provided") and when ("Use when the user asks to..." with three concrete trigger phrases). The when-clause is fully explicit with quoted triggers, matching the 5 anchor exactly; it is not the 4 anchor because the trigger guidance is not merely present but concrete and varied.

Trigger Term Quality

Opens with natural quoted phrases users would plausibly say — "measure how our narrative is landing", "track echo rate against our canon lexicon", "check how AI answer engines describe our brand" — plus "describe our brand", "echo rate", "share-of-voice". Good coverage, but common synonyms and looser variants (e.g., "brand perception", "message tracking", "are people repeating our language") are absent, and the heavy TALE/canon jargon narrows the natural-term surface — between the 4 and 5 anchors, closer to 4.

Distinctiveness Conflict Risk

Clear niche (resonance measurement only) with explicit negative boundaries naming four sibling skills: "Not for rebuilding share-of-voice machinery — use share-of-voice-tracker; not for own-site GA4/GSC analytics — use performance-monitor; not for scoring TALE profile result — use narrative-quality-auditor; not for adjudicating claims — use offer-claims-registry". Minimal conflict risk; the routing is unambiguous.