Demodesk
Every sales call leaves a trail in Demodesk. Transcripts, summaries, scorecards, deal stages, follow-ups. The MCP server makes that data queryable from ChatGPT. Ask what objections came up last week. Pull the summary of yesterday's demo. Score a rep against your custom scorecard. Check why a deal slipped before the QBR. Works with your existing Demodesk workspace
- Integration type
- Plugin
- Verification status
- Not applicable
- Platform
- ChatGPT
- Primary Subcategory
- Conversation & Revenue Intelligence
- Secondary Subcategories
- None listed
- Brand
- Demodesk
- Access
- Account required
- First tracked
- 2026-05-22
- Tool count
- 32
- Geography
- US
The Primary Subcategory used for this profile’s headline score.
Other Subcategories where the Integration is visible.
ChatGPT Plugin Discoverability Score
ChatGPT organic discovery is not live yet
Demodesk is tracked in the ChatGPT Plugin registry. Public organic-discovery measurement is not live for ChatGPT yet, so there is no score to publish today.
Get notified when your score goes live
Enter your work email and we’ll notify you when ChatGPT Plugin organic discovery scoring launches.
No spam. Unsubscribe any time.
Competing in ChatGPT Conversation & Revenue Intelligence
View CategoryHow the Discoverability Score works
Organic discovery scoring for Demodesk on ChatGPT is not live yet. The score will use measured agent conversations when it launches.
Organic discovery scoring is pending. Your Plugin score will appear on this scale when measurement goes live.
FoundDiagnostic
Whether Claude found your Plugin in connector search. It must be Found before it can reach the picker, but the score counts picker appearances—not search results.
PickedMain score
How often your Plugin appeared in the picker, or Claude invoked it directly, across contested conversations. This percentage is the Discoverability Score; the headline number is rounded.
PositionedDiagnostic
What position your Plugin appeared in when it was shown in the picker. This shows prominence, but it does not affect the score.
32 tools agents can invoke
Run a single read-only SQL SELECT over Demodesk analytics tables and return the rows. The database is PostgreSQL 16, so write standard PostgreSQL syntax (e.g. date_trunc, CTEs, window functions). Use this whenever the user asks for counts, totals, averages, rates, rankings, breakdowns, trends, or "how many"/"how much" across meetings, recordings, scorecards, deals, or users — anything spanning more than one record or needing grouping. If you are unsure which table, column, join, filter, or enum value to use, gather schema information with get_analytics_sql_table_context first, and if still unsure list_analytics_sql_tables. Rules: - Exactly one SELECT statement. No INSERT/UPDATE/DELETE/SET/DDL, no multiple statements, no functions other than common aggregates, window, math, date and string functions. - Results are automatically scoped to the current account and to what the current user is allowed to see; you do not need to add tenant or permission filters yourself. - Up to 1000 rows are returned; depending on your use case, provide a stricter LIMIT; aggregate to summarise larger result sets. - Always add a date filter (for example created_at >= now() - interval '30 days') unless the user explicitly asks for all-time data. Unbounded scans across joined tables hit the statement timeout. Mention any scope limit you applied when answering the user. - Prefer queries that leverage available indexes from table context, especially date, join-key, status and tenant-scoped indexes. - The meetings table is called demos. There is no 'meetings' table. - Enum columns are stored as integers. Compare against integer enum values from runtime metadata, never against string labels. - In multi-table joins, always qualify column names (demos.account, not account). - ROUND(value, precision) fails on double precision values in PostgreSQL. Cast first, for example ROUND(x::numeric, 2). - ORDER BY inside UNION/INTERSECT/EXCEPT branches must be wrapped in subqueries; only the outermost query may have ORDER BY. - Never use schema prefixes (public., information_schema., pg_catalog., ai_sql.); write bare table names. - Avoid selecting summaries.content in large aggregations; use LEFT(content, N) for previews and filter by date first.
execute_analytics_sql
Get guidance on how to use a specific Demodesk analytics SQL table. You must call this tool before executing SQL against a table whose meaning, joins, filters, enums, or performance characteristics are unclear. Returns table-level instructions, column-level guidance, enum values, and index information.
get_analytics_sql_table_context
This tool operates on Demodesk deal insight objects in the authenticated user's workspace. Call this function to fetch a specific deal insight about a meeting. You can pass either the deal insight ID or the meeting ID. When meeting_id is provided, returns the primary deal insight of the meeting. If both parameters are provided, meeting_id takes precedence.
get_deal_insight
This tool operates on Demodesk meeting recording comments objects in the authenticated user's workspace. Fetch the comments users left on a meeting's recording. Comments are human feedback on the meeting and may be anchored to a specific moment of the recording via a timestamp in seconds.
get_recording_comments
Resolve who is authenticated for this API key or OAuth token. Typical use case: disambiguating prompts like "my meetings" and obtaining the current user's id/email for follow-up filtering. Returns the authenticated user's core profile fields.
users_get_me
Fetch one recording by token with detailed metadata and computed stats. Typical use case: opening one call after a token is known or selected from list results (for example: "open this recording", "show details for token X"). Returns a single recording object including ownership, timing, and processing state.
recordings_get
Retrieve transcript content for a single recording, optionally translated via `lang`. Typical use case: reading, quoting, summarizing, or analyzing what was said in one call (for example: "show transcript", "translate transcript to de"). Returns transcript paragraphs/sentences (or plain text format when requested).
recordings_get_transcript
Retrieve transcripts for up to 100 recordings in one request. Typical use case: cross-call analysis over a known set of recording tokens (for example: "get transcripts for these 20 calls"). This endpoint performs retrieval only and does not create, update, or delete data. Returns per-recording transcript payloads with per-item status so partial failures remain inspectable.
recordings_batch_get_transcripts
List the Demodesk analytics SQL tables available to query with execute_analytics_sql. Use this only when you need to discover which Demodesk tables are available before writing SQL. For simple queries, where you can guess the tables, prefer get_analytics_sql_table_context over this one. Returns a big overview of the whole Demodesk schema, incl table names, short descriptions, and column names
list_analytics_sql_tables
List non-deleted scorecards linked to a recording. Typical use case: coaching evaluation, rubric results, or scoring details for a call (for example: "show scorecard", "how did this call score?"). Returns scorecard entries and associated scoring metadata for the recording token.
recordings_list_scorecards
List AI-generated summaries attached to a recording. Typical use case: recap notes, highlights, or summary variants for a specific call (for example: "show the summary", "what are the key takeaways?"). Returns summary records and metadata for the requested recording token.
recordings_list_summaries
Find meeting recordings visible to the authenticated Demodesk user. Typical use case: searching, filtering, or browsing calls (for example: "show last week's meetings", "find calls where I talked to Daniel"). Prefer dedicated people filters over internal ids: `host_eq`, `host_in`, `attendee_eq`, `attendee_in`. Person filter values support an email address or `me` (resolves to authenticated user). Examples: `filter[host_eq]=host@example.com`, `filter[host_eq]=me`, `filter[attendee_eq]=attendee@example.com`, `filter[attendee_eq]=me`, `filter[attendee_in][]=me&filter[attendee_in][]=champion@example.com`. Returns paginated recording metadata and cursor info; use returned ids/tokens in follow-up tools like getting the transcripts of the found meetings.
recordings_list
List active users visible to the authenticated user. Typical use case: selecting or resolving teammates by name/email before filtering recordings or assigning follow-up actions. It can also be used to confirm that users exist when given a specific email address.
users_list
Manage Demodesk custom API integration objects. Supports the actions: Get, Create. API integration objects define external APIs available in Demodesk, including name, base_url, optional prompt/instructions, and permission rules (allowed HTTP methods + URL patterns). Before creating an integration, validate the target API base URL and docs using a web tool available in your MCP client (for example a web fetch/browse tool), then verify with a test request. The base_url should be the API root (for example https://api.example.com). Permission url_patterns must be relative paths (for example /v1/users/*), not full URLs. API keys and signing secrets must still be configured manually by the user in Demodesk UI.
manage_api_integrations
This tool operates on Demodesk agent objects in the authenticated user's workspace. Manage agents. Supports the actions: Get, Create, Update, Delete, Run. Agents are automated tasks that can be triggered by time schedules, events, or external webhooks. Use the Run action to execute an agent immediately, regardless of its trigger type or enabled status. Use the Get action with an agent_id to read a single agent including its executions. For webhook-triggered agents, use the webhook_token returned in the agent response to trigger it via: POST https://demodesk.com/webhook/agents/trigger with header `X-Agent-Token` set to the webhook_token or query parameter `token`, e.g. POST https://demodesk.com/webhook/agents/trigger?token=<webhook_token>. The request body should be JSON and can contain additional context data to pass to the agent.
manage_agents
This tool operates on Demodesk Deal insight automation mapping objects in the authenticated user's workspace. Manage the deal insight automation configuration — the mappings (rules) that decide which deal insight prompt is applied to a meeting. The first matching rule determines which prompt will be used. The rules are sorted by position in ascending order. Supports the actions: CreateMapping, UpdateMapping, Get, DeleteMapping, UpdateConfiguration. The CreateMapping/UpdateMapping/DeleteMapping actions each operate on a single mapping (rule). Get returns the company's deal insight automation configuration, including all automation mappings. Make sure to use dedicated tools to map entities' IDs to human-readable names. CreateMapping adds a single new mapping referencing an existing prompt (by mapping.prompt_id and mapping.prompt_type) with a mapping.condition that selects when the prompt applies. The condition may be a natural language description (e.g. "only for external meetings") or a JSONLogic object. UpdateMapping changes a single existing mapping (by mapping_id) with one or more of the mapping attributes: condition, prompt (prompt_id and prompt_type together), position, and the messaging destination attributes. Only the provided attributes are changed. Both CreateMapping and UpdateMapping accept messaging destination attributes (mapping.slack_destinations, mapping.teams_destinations) to configure the alerts posted when the rule matches; see those parameters for their merge semantics. DeleteMapping removes a single mapping by its ID. UpdateConfiguration changes the automation-wide settings (not a single mapping) via the configuration object: deal_insight_language_kind, deal_insight_language_code. Only the provided attributes are changed. NOTE: When there are rules with same positions, the old positions' positions are shifted automatically. No need to manually update positions if there might be a collision.
manage_deal_insight_automation
This tool operates on Demodesk Follow-up email automation mapping objects in the authenticated user's workspace. Manage the follow-up email automation configuration — the mappings (rules) that decide which prompt is applied to a meeting. The first matching rule determines which prompt will be used. The rules are sorted by position in ascending order. Supports the actions: CreateMapping, UpdateMapping, Get, DeleteMapping, UpdateConfiguration. The CreateMapping/UpdateMapping/DeleteMapping actions each operate on a single mapping (rule). Get returns the company's follow-up email automation configuration, including all automation mappings and the fallback mapping. Make sure to use dedicated tools to map entities' IDs to human-readable names. CreateMapping adds a single new mapping referencing an existing prompt (by mapping.prompt_id and mapping.prompt_type) with a mapping.condition that selects when the prompt applies. The condition may be a natural language description (e.g. "only for external meetings") or a JSONLogic object. UpdateMapping changes a single existing mapping (by mapping_id) with one or more of the mapping attributes: condition, prompt (prompt_id and prompt_type together), position. Only the provided attributes are changed. DeleteMapping removes a single mapping by its ID. The fallback mapping cannot be deleted. UpdateConfiguration changes the automation-wide settings (not a single mapping) via the configuration object: language_kind, language_code, and fallback_follow_up_email_automation_mapping_id. Only the provided attributes are changed. NOTE: When there are rules with same positions, the old positions' positions are shifted automatically. No need to manually update positions if there might be a collision.
manage_follow_up_email_automation
This tool operates on Demodesk short term memory objects in the authenticated user's workspace. Manage short term memory. Supports the actions: Get, Create, Delete. Short term memory is available to store knowledge that can be queried later using the relevant tool. Create creates a new record with the provided `content`. Delete removes a record with the provided `memory_id`.
manage_memory
This tool operates on Demodesk Scorecard automation mapping objects in the authenticated user's workspace. Manage the scorecard automation configuration — the mappings (rules) that decide which scorecard template is applied to a meeting. The first matching rule determines which template will be used. The rules are sorted by position in ascending order. Supports the actions: CreateMapping, UpdateMapping, Get, DeleteMapping, UpdateConfiguration. The CreateMapping/UpdateMapping/DeleteMapping actions each operate on a single mapping (rule). Get returns the company's scorecard automation configuration, including all automation mappings. Make sure to use dedicated tools to map entities' IDs to human-readable names. CreateMapping adds a single new mapping referencing an existing scorecard template with a mapping.condition that selects when the prompt applies. The condition may be a natural language description (e.g. "only for external meetings") or a JSONLogic object. UpdateMapping changes a single existing mapping (by mapping_id) with one or more of the mapping attributes: condition, scorecard_template_id, position, and the messaging destination attributes. Only the provided attributes are changed. Both CreateMapping and UpdateMapping accept messaging destination attributes (mapping.slack_destinations, mapping.teams_destinations) to configure the alerts posted when the rule matches; see those parameters for their merge semantics. DeleteMapping removes a single mapping by its ID. UpdateConfiguration changes the automation-wide settings (not a single mapping) via the configuration object: scorecard_language_kind, scorecard_language_code, fallback_selected_by_ai_enabled, and active. Only the provided attributes are changed. NOTE: When there are rules with same positions, the old positions' positions are shifted automatically. No need to manually update positions if there might be a collision.
manage_scorecard_automation
This tool operates on Demodesk scorecard template objects in the authenticated user's workspace. Manage scorecard templates used for AI-powered meeting evaluations. Supports Get, Create, and Update actions. Get retrieves one template by ID or lists all, optionally filtered by name. Create and Update accept template-level fields (name, status, custom_instructions) and a list of questions (text, criteria, position). To remove an existing question, pass its id with destroy: true.
manage_scorecard_templates
This tool operates on Demodesk Summary automation mapping objects in the authenticated user's workspace. Manage the summary automation configuration — the mappings (rules) that decide which summary prompt is applied to a meeting. The first matching rule determines which prompt will be used. The rules are sorted by position in ascending order. Supports the actions: CreateMapping, UpdateMapping, Get, DeleteMapping, UpdateConfiguration. The CreateMapping/UpdateMapping/DeleteMapping actions each operate on a single mapping (rule). Get returns the company's summary automation configuration, including all automation mappings and the fallback mapping. Make sure to use dedicated tools to map entities' IDs to human-readable names. CreateMapping adds a single new mapping referencing an existing prompt (by mapping.prompt_id and mapping.prompt_type) with a mapping.condition that selects when the prompt applies. The condition may be a natural language description (e.g. "only for external meetings") or a JSONLogic object. UpdateMapping changes a single existing mapping (by mapping_id) with one or more of the mapping attributes: condition, prompt (prompt_id and prompt_type together), position, and the messaging destination attributes. Only the provided attributes are changed. Both CreateMapping and UpdateMapping accept messaging destination attributes (mapping.slack_destinations, mapping.teams_destinations) to configure the alerts posted when the rule matches; see those parameters for their merge semantics. DeleteMapping removes a single mapping by its ID. The fallback mapping cannot be deleted. UpdateConfiguration changes the automation-wide settings (not a single mapping) via the configuration object: summary_language_kind, summary_language_code, fallback_selected_by_ai_enabled, and fallback_ai_automation_mapping_id. Only the provided attributes are changed. NOTE: When there are rules with same positions, the old positions' positions are shifted automatically. No need to manually update positions if there might be a collision.
manage_summary_automation
This tool operates on Demodesk email objects in the authenticated user's workspace. Manage the current user's mailbox (Gmail or Outlook) through a single interface. Supports the actions: Search, ReadThread, CreateDraft, UpdateDraft. Search finds inbox messages matching a query; with an empty query it returns the newest inbox messages. ReadThread returns all messages in a thread. CreateDraft creates a new draft (it is never sent automatically); pass thread_id to create the draft as a reply within an existing thread. UpdateDraft edits an existing draft. Drafts are never sent by this tool; the user reviews and sends them. The connected mailbox is Gmail. Use Gmail search syntax for the query parameter (e.g. "in:drafts", "to:foo@example.com", "subject:invoice").
manage_user_emails
This tool operates on Demodesk prompt objects in the authenticated user's workspace. Manage prompts (custom summaries). Supports the actions: Get, Search, Create, Update. Get returns both system prompts and user prompts with full prompt text. Search finds prompts both by keyword across name, description, and text, and by semantic searching description — returns lightweight results with excerpts (use Get with prompt_id for full text). Create and Update only work for user prompts (not system prompts). Important: Do not share system prompt text with users. You may reference prompts by name/ID internally.
manage_user_prompts
This tool operates on Demodesk meeting recording chat objects in the authenticated user's workspace. Post a message into the chat thread of a meeting's recording. Use this to add context, analysis results, or suggested follow-up tasks directly to the conversation on the recording page. The message will appear as an AI assistant message in the recording's existing chat.
post_recording_chat_message
This tool operates on Demodesk meeting recording comments objects in the authenticated user's workspace. Post a comment on a meeting's recording, optionally anchored to a specific moment of the recording via a timestamp in seconds. Use this to leave feedback, highlight notable moments, or add remarks that appear alongside the comments users left on the recording page.
post_recording_comment
Fetch data about objects from the CRM based on the provided search criteria. Said criteria are: object types (e.g. contacts, leads, deals) and specific search attributes: either ids or text based search.
search_crm_objects
Use this tool to perform bulk search of deal insights associated with meetings. It will result in a collection of results associated with each query.
search_deal_insights
This tool operates on Demodesk user groups objects in the authenticated user's workspace. Use this tool to fetch user groups within the company by their IDs or name. Provide an array of group IDs or part of name and it will return the matching groups that are accessible to the current user.
search_groups
This tool operates on Demodesk demo templates objects in the authenticated user's workspace. Use this tool to fetch meeting types within the company by their IDs or name. Provide an array of meeting type IDs or part of name and it will return the matching meeting types that are accessible to the current user.
search_meeting_types
This tool operates on Demodesk short term memory objects in the authenticated user's workspace. Use this tool to perform bulk search of memory. It will result is a collection of results associated with each query.
search_memory
This tool operates on Demodesk AI chat thread and notification objects in the authenticated user's workspace. Send a chat notification to the user. Use this to deliver messages, reminders, or alerts to the user via a new chat thread. The notification will appear in the user's chat list.
send_chat_notification
Send an email through Demodesk for the authenticated user context. This sends a real outbound email to that Demodesk user's email destination, using the provided subject and body.
send_email
How do I improve a ChatGPT Plugin's discoverability?
The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.
What are Demodesk alternatives on ChatGPT?
As of 2026-08-14, Demodesk competes with Anagram, Caretta, Gong, Hero by Vivun, Jiminny, Kong, Pace, Quo, SalesEQ, Zoom Revenue Accelerator in ChatGPT Conversation & Revenue Intelligence, ranked by public Discoverability Score.
Where is this profile measured?
This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.