Teach your agent to use GEOly right

This week vs last: your headline metrics, the per-platform split, the competitors that moved most, new entrants, and which cited domains shifted. Formatted to paste into a channel.

Arguments

None. On a multi-brand connection, pass brand (name or id).

Output

A one-line TL;DR, your brand (mention / citation rate and sentiment with pp deltas), a per-platform table, movers up and down, a cited-domain shift table, and where to look next. Under 50 lines.

Calls

get_brand_contextget_daily_trendget_brand_overviewget_brand_boardget_citation_overview

Scan week over week for competitors that really moved: how far, which platforms they show up on, and which domains carry them while never mentioning you.

Arguments

threshold_pp — how many percentage points count as a move. Default 5.

Output

Leads with the biggest threat, then gaining competitors, losing competitors, and where your brand is most exposed.

Calls

get_brand_contextget_brand_boardget_platform_matrixget_citation_overview

Score every platform: answers, mention rate, citation rate, AIGVR, sentiment, share of mentions, and the strongest competitor there.

Arguments

None.

Output

A per-platform table plus strongest / weakest / priority platform. Priority is picked by answers × (1 − mention rate) — the pool where you are absent most. Platforms your plan covers but with no answers yet read "no data yet", never 0%.

Calls

get_brand_contextget_daily_trendget_brand_overviewget_platform_matrix

Lay AIGVR out as a topic × platform grid, calibrated to your own average so each cell reads as blind spot / weak / moderate / strong / dominant.

Arguments

None.

Output

The heatmap (up to 15 topics by answer volume), a blind-spot list, and your strongholds. Blind spots are followed up with the domains that cite competitors in that topic and not you. Cells with no data read "—" and are never called a blind spot.

Calls

get_brand_contextget_platform_matrixget_citation_overview

How your own domains perform as cited sources, which domains AI cites instead, and the content gaps where competitors are cited and you are not.

Arguments

None.

Output

Your position (share and rank, plus the you / competitor / other split), your cited domains, the content gaps, and where to act.

Calls

get_brand_contextget_citation_overviewget_domain_detail

Before/after around a launch date: how mention and citation rates moved per platform, and whether the URLs you name got picked up as AI citations.

Arguments

campaign_date — the launch date (YYYY-MM-DD, the business date the app shows); urls — comma-separated URLs to track, or omit to look at your own domains broadly.

Output

The before/after table, source pickup, and how to read it — including the reminder that this is correlation, not proof of cause.

Calls

get_brand_contextget_daily_trendget_brand_overviewget_url_detailget_citation_overview

Grade the prompt set you monitor: topic balance, tag hygiene, funnel coverage, branded vs unbranded split, duplicates, and platform data coverage.

Arguments

None.

Output

An A–F composite, a grade per dimension, and the top 3 fixes — each naming actual topics, tags, duplicate pairs or platforms. Under 10 prompts, it says the grades are not meaningful yet.

Calls

get_brand_contextget_prompt_listget_topic_list