Foglift
AI visibility data for agents
The operating layer
for agent-led AI visibility
Foglift gives an AI-native marketer, founder, or operator the data layer their agent needs: high-intent prompt checks, citation and source history, competitor context, recommendations, API access, CLI workflows, and MCP tools.
Why agents need AI visibility memory
An agent can draft pages, audits, social replies, and outreach. It still needs a reliable memory of where the brand appears, which sources AI engines cite, and what should happen next.
Agents decide the next move
A founder can ask an agent what to publish, which citation gap matters, or which newsletter deserves a pitch. Foglift supplies the prompt, citation, competitor, and recommendation data behind that decision.
AI visibility changes by prompt
One prompt may cite your homepage while another recommends a competitor. Agents need repeatable checks across ChatGPT, Perplexity, Claude, Gemini, and Google AI Overview.
Actions need source context
Generic marketing advice is not enough. The agent needs cited URLs, source history, competitor mentions, and page-level fixes so it can draft a useful edit or pitch.
How agents use Foglift
The complete agent workflow: check visibility, inspect sources, choose the move, ship the artifact.
1
Check
Agent runs a high-intent prompt across engines and sees whether the brand appears, who wins instead, and which pages were cited.
// Agent checks a category prompt
const visibility = await foglift.aiCheck({
prompt: "best AI search monitoring tool for brands",
domain: "example.com",
models: ["chatgpt", "perplexity", "claude", "gemini"]
});2
Inspect
The agent reads citation history, source URLs, competitor mentions, sentiment, and recommendation context before drafting anything.
// Agent pulls the evidence layer
const sources = await foglift.results({
days: 30,
model: "chatgpt",
domain: "example.com"
});3
Decide
Foglift recommendations turn the evidence into a ranked action list: refresh a comparison page, add a citation table, pitch a source-rich report, or adjust positioning.
// Agent asks what to do next
const recommendations = await foglift.recommendations({
domain: "example.com"
});
// Returns: prioritized pages, prompts, competitors, and rationale4
Act
The agent drafts the page edit, outreach note, social reply, or internal brief. The human reviews the artifact, then the next visibility check measures movement.
// Agent drafts from live evidence
const draft = await agent.writePitch({
sourceAsset: "/research/ai-search-citation-benchmark-2026-q2",
target: "editorial newsletter",
evidence: sources
});The growth workflow we run ourselves
Foglift's demand-gen agent uses the same loop on foglift.io. It checks high-intent AI answers, reviews cited pages and competitor mentions, filters for places where Foglift research strengthens the article, drafts founder-review outreach, and logs the result.
The human still approves what ships. The agent does the repeatable work: gather context, preserve memory, cite the source asset, and turn the next action into a reviewable draft.
Example loop
- 1. Query: test “best AI search optimization tool” across engines.
- 2. Read: collect cited URLs, repeated competitors, and missing Foglift mentions.
- 3. Match: find pages where original research answers the open question.
- 4. Draft: create a founder-review pitch tied to the canonical research URL.
- 5. Measure: rerun the prompt set after external mentions appear.
Quick Setup
Get Foglift running in your AI agent in under 60 seconds.
Option 1: MCP Server
Best for Claude Code, Cursor, and MCP-compatible agents.
npm install -g foglift-mcpAdd to your MCP config:
{
"mcpServers": {
"foglift": {
"command": "npx",
"args": ["foglift-mcp"]
}
}
}Option 2: REST API
Works with any agent that can make HTTP requests.
curl "https://foglift.io/api/v1/scan\
?url=https://example.com"Or in your agent's code:
const r = await fetch(
"https://foglift.io/api/v1/scan" +
"?url=https://example.com"
);
const data = await r.json();Designed for machine consumption
Every aspect of Foglift's output is structured for AI agents to parse, remember, and act on.
Structured JSON
Clean, typed responses with consistent schemas. No dashboard scraping required.
Citation Memory
Historical AI visibility results, cited URLs, competitors, and sentiment are queryable by an agent.
Action Ranking
Recommendations explain which page, prompt, or competitor gap deserves attention next.
Score Tracking
Numeric scores and result history give objective before-and-after measurement.
Agent-led AI visibility playbooks
These guides show how founders, marketers, and developers use Foglift data inside weekly agent workflows.
The solo-founder optimization loop
Visibility, recommendation, agent edit, repeat as the weekly operating cadence.
API-first AI monitoring
How agents query citation, sentiment, and prompt history directly from Foglift.
MCP tool evaluation
A practical lens for whether an agent can move from data to action.
AI-first content stack
The page structures that make agent-written content citable.
Visibility diagnostic loop
A 4-step workflow for finding why competitors appear when you do not.
90-day operating plan
The founder schedule for installing, measuring, and compounding visibility.
AI Discovery
Foglift publishes machine-readable documentation so AI models can discover and understand our capabilities.
llms.txt
LLM-readable site description at foglift.io/llms.txt
JSON-LD
Schema.org structured data on every page for AI understanding
MCP
Model Context Protocol server on npm
Use cases
Agent-led content planning
Have an agent inspect prompt gaps, cited sources, and competitor mentions before it drafts the next content brief.
Founder-review outreach
Turn a cited-source gap into a draft pitch that references the correct research asset and waits for human approval.
Pre-release quality gate
After an agent edits a page, scan the live URL to catch SEO and AI readiness regressions before the change ships.
AI visibility optimization
Check if a page is ready for AI search. Fix structured data, unblock AI crawlers, add FAQ sections, and rerun the prompt set.
Competitive analysis
Compare which competitors appear in AI answers, which URLs get cited, and which page format keeps winning.
Agency reporting
Agency agents can scan client sites, pull visibility history, generate reports, and propose fixes programmatically.
New: GEO Monitoring
Track AI visibility over time
Go beyond one-time scans. GEO Monitor continuously tracks how ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews mention your brand, with visibility trends, citation tracking, and sentiment analysis.
Multi-model tracking
Monitor 5+ AI search engines simultaneously. See which models mention you and which don't.
Citation tracking
Track every time an AI model links to your content. Know your citation rate and which pages get referenced.
Competitive intelligence
See when competitors appear in AI responses where you don't. Identify gaps and opportunities.
Give your agent the AI visibility layer
Let the agent read the same visibility data, citation history, and recommendations that drive the human growth workflow.