AI Visibility Monitor

Markdown delivery contract

Read markdown-delivery-contract.md for every task. Write a substantive .md document for every material work step and register its direct destination in the work index.md. For public-site work, apply the owner-required Markdown discovery profile across the full eligible inventory; preserve native standards and immutable evidence formats.

Lane outputs: visibility-baseline.md; visibility-comparison.md (combine small related steps with stable section links). Keep Markdown route fetch logs, local task completion, external retrieval traces, mentions, citations and referrals as distinct observations. Pin the same question corpus and snapshot versions for comparisons. No citation uplift inferred from deployment.

Produce a hash-pinned Visibility Run from immutable research, a formal frozen Query Corpus, repeated raw answer captures, retrieval traces where the surface exposes them, and citation reviews. This skill measures what was observed; it does not prescribe changes or claim why a metric moved. Legacy 1.0.0 runs remain readable; create new work with schema 2.0.0.

Required inputs

• Require a contract-valid Research Pack produced by ai-search-research; bind its bundle-relative path and SHA-256 into every run.
• Require a valid query-corpus.json whose Research Pack hash matches, whose frozen_at precedes every observation, and whose selected query text, locale, engine, surface, entities, and fact_ids resolve. Do not invent hidden fan-out queries.
• Require dated raw answer captures for observed cells. Preserve inaccessible, blocked, unavailable, and error states as explicit null-answer observations.
• For a comparison, require the prior Visibility Run as an immutable hash-pinned artifact.
• Read measurement-protocol.md before collecting, scoring, or comparing observations.