Reading the Output¶
Every pass produces all six deliverables — the deploy manifest whenever at least one FIX artifact was generated; an all-SHIP pass has no artifacts to record, and its confirmed-signals output is the record. Here's how to read each one.
Signal Map paragraph¶
One paragraph per dimension, in rubric order. Each paragraph leads with the finding for that dimension — no generic observations that could apply to any site.
Example from Worked Run 2:
Structured Data Coverage — Zero JSON-LD in the raw
<head>, confirmed directly from the pasted source (not a stripped fetch). No/llms.txtor/llms-full.txt. Agents parse raw HTML and guess the entity type — a brewery reads as an undifferentiated "business." This is a true, verified gap.
Score table + band¶
The five-row score table with total and band name:
| Dimension | Score (/20) | Finding |
|---|---|---|
| Structured Data Coverage | 3 | No JSON-LD (verified in raw source); no llms.txt/llms-full.txt. |
| Citability | 5 | Tagline-led homepage; facts buried under marketing. |
| Crawl Signal Clarity | 2 | robots.txt blocks all crawlers via wildcard Disallow. |
| Content Freshness | 4 | No dates, no <time>, no dateModified anywhere. |
| Entity Authority | 8 | Footer profiles present but unbound; no sameAs, no Wikidata. |
| Total | 22/100 | Band: Critical |
The band determines the next step: - Critical (0–30) — agents largely cannot find, parse, or cite this page. Fix structural issues before any content work. - Developing (31–55) — some signals present, but gaps dominate. - Functional (56–75) — solid baseline; content work can lift. - Strong (76–90) — confirmed signals only, no padding. - Authority (91–100) — all five dimensions at peak.
Per-dimension decision¶
Every dimension gets a binding SHIP / FIX / ESCALATE call from the canonical decision table:
| Dimension | Route | Why |
|---|---|---|
| Structured Data | FIX | Confirmed-absent schema + llms files; both are text artifacts. |
| Crawl Signal | FIX + ESCALATE | Corrected robots.txt is a text artifact; the Bing submission it depends on is action in the world. |
| Citability | FIX | Confirmed buried lede; rewritten opener is a text artifact. |
| Content Freshness | FIX (partial) + ESCALATE | <time>/dateModified markup is producible; wiring it to auto-update is a developer task. |
| Entity Authority | FIX (partial) + ESCALATE | sameAs is producible; Wikidata/NAP escalate. |
A dimension can also ship below 16 when no producible artifact would move its score this pass — the route is stated as SHIP (no producible gap) with the reason named, never left implicit.
Fix artifacts¶
Each FIX produces a ready-to-deploy artifact:
robots.txt replacement¶
# robots.txt
# Generated by Agent Readiness Operator fix
# -----------------------------------------------
# AI crawlers — explicitly allowed
# -----------------------------------------------
# OpenAI training crawler
User-agent: GPTBot
Allow: /
# OpenAI web retrieval crawler (ChatGPT citations)
User-agent: OAI-SearchBot
Allow: /
# Anthropic Claude crawler
User-agent: ClaudeBot
Allow: /
# Perplexity AI crawler
User-agent: PerplexityBot
Allow: /
# Google (search + AI Overviews)
User-agent: Googlebot
Allow: /
Four-part escalation¶
Every ESCALATE carries exactly four pieces:
- The dimension — which of the five.
- The specific blocker — the concrete reason the operator cannot resolve it.
- What an agent observes instead — the consequence in production.
- The exact action that unblocks it — the specific real-world step.
Example:
- Dimension: Entity Authority (Dimension 5)
- Specific blocker: No Wikidata entry exists. Creating one requires meeting notability criteria and external submission.
- What an agent observes instead: The brand is not linked into the entity graph AI models use to cross-reference, so it is cited with hedged language or omitted.
- Exact action that unblocks it: Create a Wikidata item using the facts in
/llms-full.txt, then add the resultinghttps://www.wikidata.org/wiki/Q…URL to thesameAsarray.
Deploy manifest¶
A table of all artifacts with where each goes and how to verify:
| # | Dimension | Artifact | Where it goes | How to verify |
|---|---|---|---|---|
| 1 | Crawl Signal | Corrected robots.txt |
site root /robots.txt |
curl …/robots.txt shows 5 AI bots Allow |
| 2 | Structured Data | BarOrPub JSON-LD |
<head> of homepage |
validator.schema.org → 0 errors |
| 3 | Structured Data | /llms.txt |
site root | fetch as text/plain → entity description + key pages |
| 4 | Structured Data | /llms-full.txt |
site root | fetch as text/plain → facts, not just URLs |
| 5 | Structured Data | Discovery link block | top of <body> |
links are first body element, not footer |
| 6 | Citability | Rewritten homepage opener | homepage hero | first sentence is a quotable fact, not a tagline |
| 7 | Content Freshness | dateModified + <time> |
schema block + hero | validator shows dateModified; <time datetime> renders |
| 8 | Entity Authority | sameAs (3 profiles) |
inside artifact 2 | array resolves to IG / Untappd / Google Maps |
Projected score: 22/100 (Critical) → 71/100 (Functional).