Marketing

useheading.com

Tracks brand visibility and share of voice across AI platform answers, including daily metrics, competitor leaderboards, and verbatim responses.

1 endpoint17 known toolsFirst detected October 15, 2026Last detected October 15, 2026

ENDPOINT 1

https://mcp.useheading.com/mcp

No auth detected

MCP server metadata

Name: heading
Version: 1.0.0

Capabilities

tools.listChanged

Server instructions

Heading measures how AI assistants answer questions about a brand: which prompts are tracked, how often each brand is mentioned, what the answers actually say, which sources they cite, and how the brand is depicted. Every tool except list_properties requires a propertyId. Call list_properties first to learn which ids this connection can use. Read-only. Nothing here changes any data in Heading, and only brands on a paid plan are visible. A brand that does not appear in list_properties cannot be read: asking for it by id answers "Brand not found.". All dates are UTC calendar days (YYYY-MM-DD); there are no user timezones anywhere in this data. A day range is clamped to the day the brand's daily tracking began, so a range that opens earlier comes back shorter, with a note saying why — that is not missing data.

Known tools 17

list_properties

The brands this API key can read: each one's id, name and domain.

Inferred read-only

read_visibility_summary

The brand's headline visibility numbers over a UTC day range: the overall visibility percentage (the share of tracked prompt-and-platform answers in which AI mentions this brand), the per-platform breakdown of that percentage, the count of prompts and answers measured, and the delta against the equal-length prior period where one is measurable.

Inferred read-only

read_visibility_series

The day-by-day visibility series for the brand over a UTC day range: one point per UTC day with that day's visibility percentage, optionally restricted to a single AI platform.

Inferred read-only

read_leaderboard

The share-of-voice leaderboard over a UTC day range: this brand and every tracked competitor ranked by the share of AI answers that mention each of them, with each row's mention share and rank.

Inferred read-only

read_platform_responses

The verbatim AI platform answers behind the visibility scores: for each collected response in the UTC day range, the platform, the prompt it answered, the UTC day it was fetched, whether this brand was mentioned, and the answer text itself.

Inferred read-only

read_prompts

The brand's prompt portfolio — the questions being tracked in AI platforms: every prompt with its id, exact text, status (active means tracked daily, suggested means awaiting approval, inactive means paused), quality score, latest visibility percentage, estimated search volume, and tags (branded/non-branded, intents, subjects).

Inferred read-only

read_prompt_portfolio

The brand's prompt portfolio as the composite Portfolio score measures it: the 0-100 composite score, and each of the six balance factors (tier balance, specificity, buying stage, question types, conversational tone, competitor coverage) with its score, its numeric balance target, the measured actual, which side of the target the active set sits on (over, under, or on-target — the score measures distance from the target, so overshoot is penalised exactly like undershoot), and the guidance sentence for moving toward that target.

Inferred read-only

read_prompt_changelog

The prompt portfolio change log over a UTC day range, netted to one diff per UTC day: which prompts were added to, removed from, paused in, or reactivated in the tracked set that day, plus the active-prompt count after each day's changes.

Inferred read-only

read_confidence

The brand's portfolio Confidence — how well Heading understands the business, graded from the richness of its Context documents — as last CACHED by the platform, together with each dimension's interview brief.

Inferred read-only

read_context_docs

The brand's Context tree — the markdown documents the team has curated about their business, which the platform's own pipelines (prompt generation, perception, audits) read for brand facts.

Inferred read-only

read_competitors

The brand's tracked competitors and how they compare in AI answers over a UTC day range: each competitor's name, domain, and aliases, its share of AI answers next to this brand's, and the visibility gaps — the tracked prompts where that competitor is mentioned and this brand is not, broken down by tag.

Inferred read-only

read_cited_sources

The sources AI cites when answering the tracked prompts, over a UTC day range: the domains and individual URLs that appear as citations, each with how often it is cited, its share of all citations, and whether it belongs to this brand, a competitor, or a third party.

Inferred read-only

read_perception

How AI depicts the brand over a UTC day range, measured across the perception qualities generated for this brand's category: for each quality, how often AI's answers describe the brand that way, side by side with the tracked competitors measured on the same qualities, so a quality a rival occupies and the brand is absent from is visible.

Inferred read-only

read_outcomes

The brand's outcome data over a UTC day range — the closed loop from AI visibility to business results: leads recorded through the brand's inbound lead webhook with their UTC day and attribution fields, plus the aggregate counts that tie lead flow back to the tracked visibility.

Inferred read-only

read_agent_readiness

The brand's latest Agent readiness Run — whether its own public website can be read and used by the machines that produce AI answers (ADR-0063).

Inferred read-only

read_tasks

The brand's task board and its live plans, with the ids the update tools take: every review-accepted task on a live plan or with no plan (the board's own scope) — id, title, status (backlog, todo, in_progress, done, canceled), priority, assignee (name and email), due day, plan, labels and a capped description — plus every non-archived plan with its id, title, status, linked objective and open-task count.

Potential side effects

read_research

What this brand has looked up in the index of AI answers, and the rows of one cached result.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.heading]
url = "https://mcp.useheading.com/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "heading": {
      "type": "http",
      "url": "https://mcp.useheading.com/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: heading
Remote MCP URL: https://mcp.useheading.com/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "heading": {
      "url": "https://mcp.useheading.com/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "heading": {
      "type": "http",
      "url": "https://mcp.useheading.com/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "heading",
  "transport": "streamable-http",
  "url": "https://mcp.useheading.com/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.