Report
Per-engine citation report
For each of nine engines, Hermes shows the category queries, whether you appeared, and how the answer framed you. Updated weekly.
Module 01 · Hermes Visibility
The brain. Measures where AI cites you, across nine engines, every week.
Before you write content, before you fix schema — you need to know where you stand.
Hermes measures what the engines say about you. Weekly scans across nine generative AI engines. Per-engine scoring 0–100. ShareOfAnswer aggregate. Competitor citation maps. The work list ranked by closeable gap.
Hermes is the measurement agent in the agentic workspace: it tells ContentOS and the site where to aim.
Scanned weekly
- ChatGPT
- Claude
- Perplexity
- Gemini
- Grok
- DeepSeek
- Kimi
- Google AIO
- Copilot
Open-source Lite version
Check where AI does not see you. Upgrade when you need automated scans and tracking.
AI Visibility Probe Lite is a free MIT-licensed local tool for manual AI visibility checks: brand-free discovery prompts, citation-share checks, and a Mini Briefing Test for quotability.
It runs offline with zero dependencies and no required API keys. Use Lite to test ChatGPT, Perplexity, and Gemini by hand. Move to Hermes for multi-engine scans, competitor tracking, trendlines, and weekly proof.
Lite
Free manual probe
Brand-free prompts, citation-share checks, Mini Briefing Test, MIT license, zero deps, no API keys.
Hermes
Automated visibility tracking
Scheduled scans across engines, competitor maps, ShareOfAnswer trends, source gaps, and proof packets for the workspace loop.
Why measurement is the first move
Eight in ten brands are flying blind on AI search.
AI search moved the first answer upstream.
In 2026, 800 million people ask ChatGPT a question every month. AI Overviews handle a third of Google searches.
The gap is measurable before content work starts.
Buyers who used to click your blog now read a one-paragraph AI answer. If your brand is absent, Hermes shows the gap before ContentOS starts shipping work.
Where teams go wrong
Measurement is useful only when it becomes a weekly operating loop.
Measurement without action becomes another report.
Hermes is designed to create work for the rest of the workspace. Every scan ends as a ranked list of gaps the team can close.
- Scan category questions across AI engines with buyer-language prompts.
- Compare competitors to see who gets named, cited, or ignored.
- Rank gaps by source quality, closeability, and expected citation lift.
- Send work to ContentOS so each new asset targets a measured gap.
Step 01 · Scan
Weekly per-engine query rotation
For each tier, Hermes queries every major AI engine with a rotating set of category-relevant questions. We use real buyer language, not vanity branded queries. Each scan touches ~30-100 questions per engine — enough surface area to detect ranking deltas, not just brand-name pings.
Step 02 · Score
Three-layer scoring per engine
Layer 1 = presence (does the engine mention you at all). Layer 2 = tier (first-named, second, or just enumerated). Layer 3 = evidence (does the engine attribute a specific quote or source). Aggregated to a 0-100 score per engine plus a ShareOfAnswer composite.
Step 03 · Map
Citation graph + gap list
Hermes maps which sources AI engines cite WHEN they talk about your category. That citation graph becomes the gap list: which third-party publications, review platforms, or content surfaces you are missing from. ContentOS Engine then ships work that closes the largest gaps first.
What arrives in your dashboard
The audit, the map, the work list. Re-shipped every week.
Competitors
Competitor citation map
Your top three or ten competitors get the same per-engine breakdown so the team can compare visibility side by side.
Sources
Source graph
Hermes ranks the third-party sources AI engines cite when discussing your category by citation frequency and authority weight.
Gaps
Prioritized gap analysis
The output names where rivals appear and you do not, then highlights the three moves most likely to close the gap first.
Trend
ShareOfAnswer trendline
Your aggregate visibility score over time shows whether the brand is moving closer to the default recommendation.
Export
CSV exports and shared dashboard
Teams can use read-only dashboard links and CSV pulls for BI, reporting, or follow-up analysis.
FAQ
Hermes Visibility — common questions.
How often does Hermes scan?
Weekly across Category Winner, Scale, and Done-for-you tiers. Starter gets a one-time audit. Each scan re-queries every engine with your brand + top-3 competitors across category-relevant questions; weekly cadence catches index updates that AI engines refresh every ~7-14 days.
Which 9 engines are covered?
ChatGPT, Perplexity, Gemini, Claude, Microsoft Copilot, Google AI Overviews, Yandex Neuro, Alice (Yandex), and DeepSeek. EN + RU coverage on each. We rotate the query set per category to surface citation deltas, not just brand-name pings.
What does the scoring actually measure?
Three layers: (1) presence — does the engine mention you at all on category-relevant queries; (2) tier — are you in the first-named position, the second, or just enumerated; (3) sentiment + evidence — does the engine attribute a specific claim or quote a source. Score is normalized 0-100 per engine, plus an aggregate ShareOfAnswer.
Can I track competitors?
Yes — Starter tracks you only; Category Winner tracks you + 3 competitors; Scale tracks you + 10 competitors; Done-for-you is custom. The citation map shows who AI engines cite WHEN talking about your category — so you see exactly which sources to displace.
How is Hermes different from Profound, Otterly, AthenaHQ?
Hermes is part of a closed loop with ContentOS (production) + Website Agentic Optimization (foundation). Profound/Otterly/AthenaHQ are measurement-only — they tell you the score, you go fix it elsewhere. The 4-tool stack costs $1.5-3K/mo. Humanswith.AI Scale at $2,497/mo replaces all of them with one closed-loop platform.
Is the data exportable?
Yes — CSV export of citation events + per-engine scores. Hermes also has a read-only dashboard URL you can share with your team. Webhook integration to Slack/Telegram/Twenty CRM available on Scale+ tiers.
Run your brand through Hermes before the call.
30-minute strategy call. The audit lands in your inbox first.
You arrive to a per-engine citation map of your category, the closeable gaps, and an honest read on whether any tier fits. No deck. No pitch. Just the audit.