Model Context Protocol (MCP) tools let AI clients use your Pendo data, including metadata, product usage, guide performance, and feedback, to answer product questions. Tools return structured responses that are safe for automated workflows and agents like Claude or ChatGPT. While most tools are read-only, some allow you to write data back to Pendo, such as submitting feedback.
To connect your AI client to Pendo, follow the steps in Connect to the Pendo MCP server.
This article explains what’s available in MCP by product area (which parts of Pendo agents can access) and by tool (what each one does, how to prompt it, and key limits).
Note: Metadata schema tools show the available fields for use in queries. Query tools return account or visitor data based on those fields. The fields included in query results depend on your subscription’s available metadata and how your AI client constructs its request.
Some tools are used together to answer more complex questions. For example, list_all_applications can supply Application IDs for use in activityQuery, and visitorMetadataSchema helps identify which metadata fields to include when constructing usage queries.
What you can access by product area
The table below shows which Pendo product areas MCP can access and which tools power each area. Use it to confirm whether an agent can see a given type of data before you design prompts or workflows around it.
| Product area | What you can access by MCP | Key tools |
| Pages and Features | Usage metrics, including events, visitors, and accounts |
activityQuery, productAreaMemberActivity, searchEntities, listProductAreas
|
| Visitors and accounts | Metadata and counts, last activity, segment filtering |
visitorQuery, visitorMetadataSchema, accountQuery, accountMetadataSchema
|
| Guides and polls | Guide-level metrics, including views, completions, dismissals, poll activity, and NPS metrics |
activityQuery, guideMetrics, searchEntities, listGuides, npsScore
|
| Session Replay | Replay metadata and links with filters; frustration events, developer console logs, and network events |
sessionReplayList, devlogEvents
|
| Listen | Raw feedback, ideas, insights, clustered themes, and submitting feedback |
get_feedback_items, get_feedback_insights, generate_feedback_topics, get_ideas, createFeedbackItem
|
| Agent Analytics | Key metrics, AI agents, use cases, issues, and issue diagnosis |
agent_analytics_key_metrics, list_ai_agents, list_use_cases, list_ai_agent_issues, ai_agent_issue_analysis
|
Available MCP tools
The table below lists each externally available MCP tool, what it’s for, example prompts, and important limits like time ranges and result caps. Use it as a reference when choosing tools or debugging why a question does or doesn’t return data.
| Tool | What it's for | Example prompts | Why it's useful | Limits and availability |
list_all_applications |
Retrieve a list of subscriptions and applications you have access to. Often used as a first step with other tools (for example, passing App IDs to Also confirms which application data you have access to through authentication. |
List all apps in my subscription. Which apps belong to the Acme Co subscription? Use all App IDs to check top features by usage. |
Provides the Subscription IDs and Application IDs needed for other tools. | Basic subscription and application information only; no subscription settings. |
accountMetadataSchema |
View the available metadata fields for accounts, such as ARR, industry, renewal date, or product tier. Returns field names, types (for example, boolean or string), and a historical flag indicating whether the field supports historical (event-time) filtering. |
What metadata fields are available for accounts? Can I filter accounts by ARR or renewal date? What account attributes exist in our dataset? |
Shows which account fields you can include in queries or use to filter data. | N/A |
accountQuery |
Retrieve account records that match specific criteria based on available metadata fields. |
List all accounts with ARR over 100k. Which accounts are up for renewal next month? How many accounts are in the @Enterprise segment? Which accounts haven't signed in over the last 30 days? |
Returns account data so you can analyze usage, renewal risk, or growth opportunities. | N/A |
visitorMetadataSchema |
View the available metadata fields for visitors, such as job title, CRM fields, region, or role. Returns field names, types (for example, boolean or string), and a historical flag indicating whether the field supports historical (event-time) filtering. |
What metadata fields are available for visitors? List every Salesforce field we have at the visitor level. Which visitor fields are available for analyzing usage by role or onboarding status? |
Helps you identify which visitor attributes you can use when filtering or organizing usage data. | N/A |
visitorQuery |
Retrieve visitor records that match specific criteria based on available metadata fields. |
List all visitors whose last page view was before June 1, and include their browser and country. How many users are labeled as “beta tester” across all apps? |
Returns visitor data so you can analyze user behavior and engagement trends. |
Maximum of 50,000 visitors for each query. Anonymous visitors excluded by default. Results sorted by last visit time, most recent first. |
activityQuery |
Retrieve product usage metrics for Pages, Features, Track Events, accounts, visitors, and polls. Supports frustration metrics (error clicks, rage clicks, U-turns, dead clicks) when applicable. Supports Product Area filtering to narrow queries to specific product areas. |
How many unique users interacted with our new onboarding flow this month? Which Features have the lowest engagement? What are the top 20 most visited pages in the last month? Show me the most and least used features in the last 30 days. What are the top pages in product area X? |
Provides usage metrics to support product decisions, identify at-risk accounts, and measure the impact of new releases. |
Maximum date range: 367 days. Returns aggregated metrics only; not individual events. Supports day and week ranges (cumulative) or daily and weekly periods (iterative breakdowns). |
productAreaMemberActivity |
Return all Pages, Features, and Track Events belonging to a product area, including those with zero activity in the date range. |
Which features in the Onboarding product area have zero usage this quarter? What pages in the Analytics product area should we consider sunsetting? |
Helps you identify unused or low-engagement entities within a product area. | Maximum date range: 367 days. |
listProductAreas |
List all product areas for a subscription, including their IDs, names, and descriptions. Supports optional fuzzy search and pagination. |
What product areas do we have? List all product areas for this subscription. Show me all available product areas. |
Helps you enumerate product areas or find a Product Area ID to pass to other tools that require it. | N/A |
guideMetrics |
Returns performance data for guides. |
How did our onboarding walkthrough perform this month? What percentage of users finished the @Setup Checklist guide? Show the dismissal rate trend for the @Trial Expiration Warning guide this quarter. |
Provides onboarding, adoption, and customer experience metrics for launched guides. |
Maximum date range: 367 days. No step-level metrics or button-level clicks. Supports range (cumulative) or iterative (daily/weekly) date ranges for core guide metrics. (Poll, NPS, and adoption metrics like total viewers, converted visitors, and goal adoption rate require a date range). |
listGuides |
List, filter, and optionally retrieve full content for in-app guides, including IDs, names, states, types, and scheduling. |
List all public guides. Show me draft guides. Which guides are set to launch automatically? List guides that have expired. Show me all tooltip guides that are currently staged. What does the "Onboarding tour" guide say to users? |
Helps you find guides by status, type, activation method, or other attributes, and review full content for specific guides when needed. | Supports search on guide names and descriptions using semantic or "fuzzy" matching. Can return either summary lists of guides or full content for a single guide; full content is available for eligible guides only. |
npsScore |
Return NPS (Net Promoter Score) metrics for NPS surveys, including the overall NPS score (-100 to +100), average rating, and response distribution across promoters, passives, and detractors. |
What is our NPS score? Show me NPS trends over the last 30 days. What percentage of respondents are detractors? |
Provides an overview of customer sentiment and survey response breakdown without requiring a custom report. | Maximum date range: 367 days. Can optionally return up to 500 individual responses if a guide ID is provided. |
searchEntities |
Find Pendo items by name or conceptual meaning, including Pages, Features, Track Events, accounts, and saved clips. (Deprecated for guides and Product Areas).
|
List all Pages that contain “/settings” in the URL pattern. Find every Feature that includes “export” in its name. |
Helps you locate the Pages, Features, and events needed for building queries or exploring product usage. | Results ranked by relevance. Uses "fuzzy" matching when semantic similarity is low. |
segmentList |
List all publicly shared segments with their names and IDs. |
List all the segments that contain “current customers” in the name. Which segment includes at-risk enterprise accounts? |
Helps you identify available segments for use in other queries. | N/A |
productEngagementScore |
View product health using Pendo’s Product Engagement Score (PES) for adoption, stickiness, and growth. Supports custom configurations for user base, stickiness calculations, and event filters. |
What’s our PES for the last 30 days? Calculate adoption for our top five Features. What’s our stickiness score for accounts, weekly over monthly? |
Provides adoption, stickiness, and growth metrics for a complete view of product health and engagement. | Maximum date range: 180 days. |
list_ai_agents |
Retrieve a list of all AI agents you have access to. Often used as the first step to get Agent IDs for use with other Agent Analytics tools. |
List all AI agents in my subscription. What agents are configured for my app? |
Provides the agent IDs needed to run queries. | N/A |
agent_analytics_key_metrics |
Retrieve key aggregate metrics for an AI agent's conversations, including prompt volume, visitor and account counts, rage prompt rates, and retention. |
How many conversations has my support agent had in the last 30 days? What's the rage prompt rate for our onboarding agent this month? Show me key metrics and trends for my chat agent over the past two weeks. |
Provides a high-level summary of agent performance and engagement trends without requiring you to build a custom report. | Maximum date range: 90 days. |
list_use_cases |
Retrieve conversation clustering analysis for an AI agent, grouping prompts and conversations by semantic topic. |
What topics are visitors asking our support agent about most? Show me the top use cases for our onboarding agent last quarter. |
Helps you understand what visitors are actually trying to do with your AI agent, without reading individual conversations. | Maximum date range: 90 days. Requires an App ID and an Agent ID from list_ai_agents. |
list_ai_agent_issues |
Retrieve detected issues in AI agent conversations. |
What problems are users running into with our support agent? Show me the most common issues in our chat agent over the last 30 days. Have we seen fewer incorrect answer issues since last month? |
Surfaces patterns of failure or frustration in AI agent conversations so you can prioritize improvements. | Maximum date range: 90 days. Limited to 500 issues; narrow by Agent ID or date range if needed. |
ai_agent_issue_analysis |
Retrieve issue diagnoses, flagged response tool and model usage, and visitor prompt content for the events of a specific detected issue cluster. |
What patterns exist among instances of the "incorrect answers" issue? Which tools were called in instances of the "timeout error" issue? Show me what users said in conversations flagged as the "authentication failure" issue. |
Helps you understand why conversations were flagged, see patterns in visitor prompts, and identify which tools and models were involved. | Maximum date range: 90 days. Requires an Agent ID, conversation IDs, and event IDs from list_ai_agent_issues. |
sessionReplayList |
Returns links to replays for specific visitors or behaviors. Default filters: minimum duration 120 seconds, minimum activity 5%. Exclude list behavior options available. Supports event property filtering. |
Show me replays from the last week with high activity. Show me sessions with frustration events in the onboarding flow. Show me replays where users visited the error page. Show me replays where the checkout-completed Track Event occurred. Show me replays for the Onboarding product area. |
Helps you find high-value sessions to review, making it easier to identify friction or unexpected user behavior. |
Maximum of 50 replays for each query. Maximum date range: 31 days. Replay content not exposed through the MCP. |
devlogEvents |
Retrieve raw development logs for a visitor or session, including HTTP request/response details, log levels, messages, and stack traces. Accepts a clip ID or the individual fields that identify a recording to resolve context automatically. If you have a session replay URL, extract the required parameters from it before calling this tool. |
Show me devlogs for this replay: [link] Show network requests captured in clip [link] What HTTP errors occurred during visitor john's session? Get devlogs for clip ID abc-123 Find devlog errors for account acme-corp What API calls did this visitor make during their session? |
Helps you investigate developer console logs, network errors, or HTTP activity captured during a replay. |
Maximum of 100 events for each query. Maximum date range: 367 days. |
get_feedback_items |
Retrieve raw customer feedback items that match specific filters, including the original title, description, and creation date, along with associated account and visitor details. Filter by account types, alerts, feedback types, Product Areas, and more. |
Show all feedback items from enterprise accounts about onboarding. List raw feedback for accounts in the @At-risk segment from the last 30 days. Get all feedback where the title or description mentions “performance”. |
Returns the full, unprocessed feedback so you can review exact customer language and context without relying on summaries or extracted insights. |
Maximum of 30 feedback items for each call. Raw feedback only. |
get_feedback_insights |
Retrieve extracted feedback insights from customer feedback that match specific filters, including summaries, explanations, and supporting quotes. Filter by account types, alerts, feedback types, Product Areas, and more. |
Summarize the main feedback insights from customers on our new reporting page. Show feedback insights about reliability issues from the last quarter. List top insights from feedback submitted by enterprise customers. |
Helps you quickly understand key themes and takeaways from customer feedback without reading every individual feedback item. |
Summaries and topics only, not full dataset. May require multiple calls for large date ranges. |
get_ideas |
Retrieve product ideas that match specific filters. Each item includes a title and description. Supports the same filters as other feedback tools, including account types, Product Areas, date range, and search terms. |
What ideas have been submitted about our onboarding experience? Show me ideas about analytics submitted in the last quarter. |
Helps you quickly surface and review product ideas alongside feedback without leaving your AI client. | Maximum of 30 items for each call. |
generate_feedback_topics |
Generate AI-clustered topics from customer feedback that matches specific filters, with each topic including a name, description, and count of related insights. Filter by account types, alerts, feedback types, Product Areas, and more. |
Generate topics from all feedback about onboarding in the last 90 days. Show the main themes in feedback from customers who recently downgraded. Cluster feedback about performance issues into topics. |
Provides a high-level view of the most common feedback themes to support prioritization and trend analysis. |
Summaries and topics only, not full dataset. May require multiple calls for large date ranges. |
createFeedbackItem |
Create a new feedback item in Listen. |
Log a feature request for dark mode support. Submit feedback that the export CSV button is confusing. Capture this bug report about broken pagination. |
Allows you to submit product feedback, feature requests, or bug reports directly from your AI client. | Requires an App ID, title, description, and at least one Visitor ID or Account ID. |