Agent-First GEO Intelligence Platform

Market intelligence built for AI agents

MentionMap is the research backend for autonomous agents. Agents onboard products, trigger research across LLM ecosystems, and get structured intelligence to improve AI recommendation visibility — via API, CLI, or Agent Skill.

pip install mentionmap-agent-skill See the Agent Workflow →

Researches across

ChatGPTClaudePerplexityGemini

Agents are the primary operators

MentionMap is designed for autonomous AI systems. Every operation is an API call. Every response is structured JSON. Humans get a dashboard. Agents get a skill.

OLD WAY

→ Marketer logs into dashboard

→ Clicks through 5 screens manually

→ Exports CSV to analyze

→ Writes a report for the team

→ Manually updates content strategy

AGENT-FIRST

→ Agent calls create_product

→ Agent calls trigger_research

→ Agent polls check_research_status

→ Agent reads get_recommendations

→ Agent acts on recommendations autonomously

End-to-end agent workflow

Your agent launches a product. MentionMap gives it market intelligence. It acts.

from mentionmap_skill import MentionMapSkill

mm = MentionMapSkill(api_key="...")

# 1. Onboard
company = mm.create_company("Vibe Engineering", domain="vibeeng.com")
brand = mm.create_brand(company.id, "Agent Docs")
product = mm.create_product(brand.id, "AgentDocs", category="ai-docs")

# 2. Research (async — agent continues other work)
job = mm.trigger_research(product.id)

# 3. Get intelligence
competitors = mm.get_competitors(product.id)
share = mm.get_recommendation_share(product.id)
actions = mm.get_recommendations(product.id)

# 4. Act on recommendations
for action in actions:
    if action.type == "create_content":
        agent.create_content(action.topic, action.target_source)

How MentionMap works

Five steps from onboarding to action

Agent Onboards

Your agent registers a company, brand, and product via API, CLI, or Agent Skill. Structured inputs, no forms.

Research Runs

MentionMap executes structured prompts across ChatGPT, Claude, Perplexity, and Gemini — simulating real buyer questions.

Intelligence Extracted

Recommendations, citations, competitors, strengths, weaknesses, and narratives are extracted from every response.

Scores Computed

Recommendation Share, Citation Share, and Source Influence scores are calculated and tracked over time.

Actions Generated

Evidence-backed GEO actions are prioritized: content to create, pages to improve, sources to target.

Three ways to integrate

Agent Skill, CLI, or direct API. All return structured JSON.

Agent Skill

Strongly-typed Python client for AI agents. Compatible with OpenClaw, LangGraph, LangChain, and MCP.

pip install mentionmap-agent-skill

CLI

Scriptable command-line interface. Every operation as a command. JSON output for machine consumption. CI/CD friendly.

pip install mentionmap-cli

REST API

Versioned JSON API at /api/v1/. Bearer token auth. Async research with job polling. Webhook support.

api.mentionmap.ai/api/v1

12 tools. Structured inputs. Structured outputs.

Every operation your agent needs

create_company

create_brand

Create a brand under a company

create_product

trigger_research

Run research across AI models

check_research_status

Poll async research status

get_competitors

Discover AI-recommended competitors

get_citations

See which sources AI cites

get_sources

Rank sources by influence score

get_recommendation_share

Your AI recommendation score

get_recommendations

Prioritized GEO actions

ask_strategy

Data-grounded strategy Q&A

register_assets

Link docs and URLs

CLI for humans and machines

Scriptable. CI/CD friendly. JSON output.

$ mentionmap product create AgentDocs \
    --brand agent-docs \
    --category ai-documentation \
    --domain agentdocs.ai
✓ Created product: AgentDocs

$ mentionmap research start agent-docs
⟳ Research job submitted: job_a1b2c3d4

$ mentionmap research results agent-docs --format json
{
  "recommendation_share": 0.34,
  "competitors": ["Docusaurus", "GitBook", "ReadMe"],
  "top_action": "Create comparison page vs Docusaurus"
}

Built for every operator

From autonomous agents to marketing dashboards

Autonomous Agents

OpenClaw, LangGraph, or custom agents use MentionMap as their market intelligence backend.

Technical Founders

Integrate GEO research into your product pipeline. Know what AI says about you programmatically.

Marketing Teams

Dashboard view of AI visibility. Recommendation Share scores. Weekly trend reports.

DevOps & CI/CD

Run GEO audits in CI. Block releases if recommendation share drops. Automate positioning checks.

Get Your Free AI Visibility Audit

Enter your product and category. We'll show you your Recommendation Share, competitor mentions, and top GEO actions.

Questions & Answers

Give your agents market intelligence.

The research backend for autonomous AI operators.

Get Your Free Audit View on GitHub