How is this different from connecting Smartsheet MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Smartsheet MCP API changes?

We track the Smartsheet MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Smartsheet tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Smartsheet?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Smartsheet MCP server myself?

No. TrustGate connects to the Smartsheet MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Smartsheet directly.

Connect Smartsheet MCP to your AI agents, securely.

Q: Is Smartsheet MCP secure?

The server itself exposes 83 tools, and some of them can write to your Smartsheet account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Smartsheet.

One secured endpoint for every Smartsheet call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Smartsheet MCP now

What your agents can ask once Smartsheet MCP is connected.

Select the tools your AI agents can call

83 tools from the Smartsheet MCP schema.

Showing 83 tools

searchreadSearches across all accessible Smartsheet assets by name or content. Scopes cover sheet, report, dashboard, folder and workspace names, cell data, comments, attachments, summary fields and template names. Returns up to 100 results with no pagination. Scope: READ_SHEETS.
browse_workspacereadExplores a workspace's top-level contents, filterable by folders, sheets, reports, dashboards and templates, with token-based pagination up to 1000 items per page. Scope: READ_SHEETS.
browse_folderreadNavigates a folder's direct children, returning nested folders, sheets, reports and dashboards with IDs, names, permalinks and ownership. Scope: READ_SHEETS.
list_workspacesreadLists the workspaces in the account with cursor-based pagination. Scope: READ_SHEETS.
list_sharesreadLists the groups and users an asset is shared with.
create_workspacewriteCreates a new workspace, the top-level container for sheets, reports, dashboards and folders. Names are capped at 50 characters. Scope: ADMIN_WORKSPACES.
create_folderwriteCreates a folder inside a workspace or parent folder. Names must be unique within the parent and the caller needs ADMIN permissions on it. Scope: ADMIN_WORKSPACES.
create_workspace_assetshigh riskAn AI agent that builds end-to-end Smartsheet solutions from a natural language description, creating any combination of workspaces, sheets, dashboards, workflows, reports, forms, portfolios, projects and blueprints in a single call. Scopes: ADMIN_WORKSPACES, ADMIN_SHEETS, CREATE_SHEETS, WRITE_SHEETS, CREATE_SIGHTS.
get_sheet_pathreadGets a sheet's full hierarchical path and a direct permalink without loading the sheet itself. Scope: READ_SHEETS.
get_sheet_summaryreadRetrieves sheet data with multi-column filtering (AND logic only), sorting and column selection. Limited to roughly 8,000 tokens, so very large sheets return sampled rows. Scope: READ_SHEETS.
get_sheet_versionreadReturns a sheet's current version number without loading it. The version increments on cell changes, comments, row modifications, attachments and formatting updates. Scope: READ_SHEETS.
find_in_sheetreadSearches for text within a sheet, returning matching cells with row and column IDs, display values and character offsets. Supports case sensitivity, column scoping, formula pre-filtering, sorting and pagination up to 20,000 results. Scope: READ_SHEETS.
create_sheetwriteCreates a sheet in a folder or workspace, either blank with a custom column schema or from a template with selective element copying. Supports Gantt, dependency and resource management configuration. Scope: CREATE_SHEETS.
get_columnsreadRetrieves column metadata: IDs, names, types and properties such as picklist options and system column designations. Scope: READ_SHEETS.
add_columnswriteAdds one or more columns with type, position, formula, validation, locking, auto-numbering and tags. Supports AISUMMARIZE, AITRANSLATE and AISENTIMENT formulas. Scope: ADMIN_SHEETS.
update_columnhigh riskModifies an existing column's title, type, position, visibility, formula, picklist options and validation. Marked destructive and irreversible: changing a column type can cause data loss or unexpected conversions across every row. Scope: ADMIN_SHEETS.
delete_columnhigh riskPermanently removes a column and all its data from every row. Marked destructive and irreversible, with no rollback. Scope: ADMIN_SHEETS.
get_cell_historyreadRetrieves a cell's complete modification history with timestamps, old and new values and user attribution. Rate limited to 30 requests a minute. Scope: READ_SHEETS.
add_rowswriteAdds rows with location specifiers for position and hierarchy, up to 500 per call. Cell values over 4000 characters are silently truncated. Scope: WRITE_SHEETS.
update_rowshigh riskModifies existing row data and position, up to 500 rows per call. Marked destructive and irreversible, with no automatic undo. Scope: WRITE_SHEETS.
delete_rowshigh riskPermanently removes rows and every cell value in them, up to 10 per request. Marked destructive and irreversible. Scope: WRITE_SHEETS.
list_attachmentsreadLists attachments at sheet, row or discussion level, covering files and Google Drive, OneDrive, Box and Dropbox links. Scope: READ_SHEETS.
get_attachmentreadRetrieves attachment metadata including name, type, size, timestamps and a temporary download URL that expires. Scope: READ_SHEETS.
attach_urlwriteAttaches a URL to a sheet, row or comment, typed as a generic link or as Google Drive, Box, Dropbox, Evernote, Egnyte or OneDrive. The URL itself is not validated. Scope: WRITE_SHEETS.
delete_attachmenthigh riskRemoves an attachment from a sheet. Marked destructive and irreversible; deleted attachments cannot be recovered through the API. Scope: WRITE_SHEETS.
list_discussionsreadRetrieves discussions across a sheet or on a specific row, optionally including comments and attachments. Scope: READ_SHEETS.
get_discussionreadRetrieves a single discussion by ID with its comments and attachments included automatically. Scope: READ_SHEETS.
create_discussionwriteStarts a discussion thread on a sheet or row. The title is generated from the first 100 characters of the opening comment. Scope: WRITE_SHEETS.
add_commentwriteAdds a reply to an existing discussion thread. Standalone comments cannot be created. Scope: WRITE_SHEETS.
update_commenthigh riskReplaces the entire text of an existing comment. Marked destructive. Only the comment's author can update it. Scope: WRITE_SHEETS.
delete_commenthigh riskPermanently removes a single comment. Marked destructive and irreversible. Only the author can delete it. Scope: WRITE_SHEETS.
delete_discussionhigh riskPermanently removes an entire discussion thread and all of its comments. Marked destructive and irreversible. Scope: WRITE_SHEETS.
list_reportsreadLists the reports the user can access with their ID, name, access level and permalink. Scope: READ_SHEETS.
get_reportreadRetrieves a report's columns, rows and cell values with pagination up to 10,000 rows per page, optionally including attachments, discussions, format and source sheets. Scope: READ_SHEETS.
create_reportwriteCreates a row or sheet summary report aggregating up to 100 source sheets or workspaces, with up to 400 columns, filters nested three levels deep, grouping, sorting and SUM, AVG, MIN, MAX, COUNT, FIRST and LAST summarizing. Scope: READ_SHEETS.
get_workflowsreadLists a sheet's workflows with each one's full trigger and action tree, in the same format create_workflow accepts. Reminder-type workflows are not returned. Scope: READ_SHEETS.
create_workflowwriteCreates a workflow from a typed tree using column titles and recipient emails rather than IDs. Enabled by default on creation. Scope: WRITE_SHEETS.
enable_workflowwriteRe-activates a disabled workflow. Idempotent. Scope: WRITE_SHEETS.
disable_workflowwriteTurns off a workflow without deleting it, notifying the sheet's admins. Reversible and idempotent. Scope: WRITE_SHEETS.
update_workflowhigh riskFully replaces a workflow's definition with the tree sent; anything omitted is removed. Marked destructive and irreversible, and it invalidates in-flight approvals so pending approval emails stop being actionable. Supports an optimistic-concurrency version guard. Scope: WRITE_SHEETS.
delete_workflowhigh riskPermanently deletes a workflow. Marked destructive and irreversible; it cannot be recovered or re-enabled. Scope: WRITE_SHEETS.
list_dashboardsreadLists accessible dashboards with ID, name, access level and timestamps, using token-based pagination. Scope: READ_SIGHTS.
get_dashboardreadRetrieves a dashboard with its widget configuration, layout and optional source and email address data. Scope: READ_SIGHTS.
create_dashboardwriteCreates a dashboard whose widgets are AI-generated from up to five source sheets and a natural-language prompt. Rolls back automatically if creation fails. Scope: CREATE_SIGHTS.
get_chart_suggestionsreadReturns AI-powered chart suggestions for a sheet, grouped into up to five categories with up to four suggestions each, including ready-to-use prompts. Scope: READ_SHEETS.
create_chart_from_promptreadGenerates charts and metrics from a natural-language description of a sheet's data, supporting multi-turn conversations. The data is returned for the client to render rather than rendered automatically. Scope: READ_SHEETS.
create_chart_from_datareadBuilds a bar, line, pie or stacked bar chart structure from pre-computed values without querying any sheet. Requires no access scope.
refine_chartreadRefines a chart produced by create_chart_from_prompt, using that result's conversation and message IDs.
explain_chartreadProduces a plain-language summary of how a generated chart or metric was derived from the underlying data.
add_charts_to_dashboardwriteAdds charts generated in a chart conversation to a dashboard.
query_chartreadQueries a saved chart's data, returning the definition, data points and styling needed to render it. Scope: READ_SHEETS.
get_chart_definitionreadRetrieves a saved chart definition, including its type, axes, styling and configuration. Scope: READ_SHEETS.
list_favoritesreadLists all items marked as favorites by the current user. Scope: READ_SHEETS.
is_favoritereadChecks whether a specific asset is marked as a favorite by the current user. Scope: READ_SHEETS.
add_favoriteswriteMarks one or more assets as favorites for the current user. Scope: ADMIN_WORKSPACES.
delete_favoriteswriteRemoves one or more items from the current user's favorites. Scope: ADMIN_WORKSPACES.
list_portfoliosreadLists the portfolios accessible to the current user, with name, workspace ID and permalink. Scope: READ_SHEETS.
create_portfoliowriteCreates a new portfolio. Scope: ADMIN_WORKSPACES.
get_portfolioreadGets a portfolio by ID, with its name, workspace ID and permalink. Scope: READ_SHEETS.
delete_portfoliohigh riskPermanently deletes a portfolio, including all its project template and project workspaces and their assets. Scope: ADMIN_WORKSPACES.
list_project_templatesreadLists the project templates in a portfolio. Scope: READ_SHEETS.
create_project_templatewriteCreates a new project template within a portfolio. Scope: ADMIN_WORKSPACES.
get_project_templatereadGets a project template by ID, with a summary of its parent portfolio. Scope: READ_SHEETS.
delete_project_templatehigh riskPermanently deletes a project template, including all assets in its workspace. Scope: ADMIN_WORKSPACES.
list_projectsreadLists the projects created from a project template. Scope: READ_SHEETS.
get_projectreadGets a project by ID, with its name, workspace ID and permalink. Scope: READ_SHEETS.
get_portfolio_project_list_sheetreadRetrieves the project list sheet of a portfolio. Scope: READ_SHEETS.
get_scenario_groupsreadLists scenario plans from Resource Management, with pagination.
get_scenario_groupreadGets a scenario plan by ID from Resource Management.
create_scenario_groupwriteCreates a new scenario plan in a workspace.
delete_scenario_grouphigh riskPermanently deletes a scenario plan and all its scenarios.
create_scenariowriteAdds a new what-if scenario to an existing scenario plan.
get_scenario_diffsreadReturns the cell, row and column differences between a scenario and its source sheets, including conflicts.
get_scenario_capacity_workloadreadReturns capacity and workload hours for a scenario over a date range, grouped by user, discipline or role.
apply_scenariohigh riskApplies a scenario's changes to its source sheets.
get_shadow_sheetreadRetrieves a scenario sheet, the scenario-specific copy of a source sheet, with its columns and rows.
get_shadow_sheet_rowreadRetrieves a single row from a scenario sheet.
add_shadow_sheet_rowswriteAdds one or more rows to a scenario sheet.
update_shadow_sheet_rowswriteUpdates one or more rows in a scenario sheet.
delete_shadow_sheet_rowshigh riskPermanently removes one or more rows from a scenario sheet.
get_plan_idreadLists the plans the authenticated user is a member of, to find a plan ID. System admins only.
get_seat_usagereadRetrieves per-user seat usage and activity scores for a plan. Scope: ADMIN_USERS.
get_license_inventoryreadRetrieves a point-in-time license usage snapshot for a plan. Scope: ADMIN_USERS.

What Smartsheet MCP exposes

Smartsheet MCP groups its 83 tools into 16 categories, each with a suggested starting policy.

All 83 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Discussions & Comments	Listing and reading discussions at sheet or row level, creating and deleting discussion threads, and adding, updating and deleting comments	Read / Write	Deleting a discussion removes the whole thread and all its comments. Only the author can update or delete their own comment
Automated Workflows	Listing a sheet's workflows with their full trigger and action trees, creating one from a typed tree, replacing it, enabling and disabling it, and deleting it	Read / Write	update_workflow fully replaces the stored tree, so anything omitted is removed, and it invalidates in-flight approvals: pending approval emails stop being actionable. Prefer disable_workflow over delete_workflow
Charts	AI chart suggestions for a sheet, creating charts from a natural-language prompt or from pre-computed values, refining and explaining them, and adding them to a dashboard	Read / Write	Chart data is not rendered automatically; the client renders it. Only sheets work as data sources, not reports or dashboards
Workspaces & Folders	Browsing workspace and folder contents with type filtering and pagination, listing workspaces and an asset's shares, and creating workspaces and folders	Read / Write	Creation needs ADMIN_WORKSPACES. list_shares reveals who an asset is shared with, which is useful for an access review and sensitive on its own
Sheets	A sheet's hierarchical path and permalink, filtered and sorted sheet data, the current version number for change detection, text search within a sheet, and sheet creation from scratch or a template	Read / Write	get_sheet_summary caps at roughly 8,000 tokens and samples rows beyond that, so a result is not necessarily complete. get_sheet_version is the cheap way to poll for changes
Columns	Reading column metadata and types, adding columns with formulas, validation and system types, and updating or deleting a column	Read / Write	Both update_column and delete_column are marked destructive and irreversible. Changing a column type can cause data loss across every row, and deleting one removes its data from the whole sheet
Rows	Adding rows with hierarchical positioning, updating existing rows, deleting rows and reading a cell's full modification history	Read / Write	update_rows and delete_rows are marked destructive and irreversible. Deletion is capped at 10 rows per request, adding and updating at 500. get_cell_history is rate limited to 30 requests a minute
Attachments	Listing attachments at sheet, row or discussion level, reading attachment metadata with a temporary download URL, attaching a URL and deleting an attachment	Read / Write	attach_url performs no validation of the URL, so an agent can attach any string. delete_attachment is irreversible and the file cannot be recovered through the API
Reports	Listing accessible reports, retrieving a report's columns, rows and cell values with pagination, and creating a report aggregating up to 100 sheets or workspaces	Read / Write	Reports are read-only views of source sheets. A created report can aggregate across workspaces, so it widens what one asset exposes
Dashboards	Listing dashboards with their access levels, retrieving one with its widget configuration and layout, and creating a dashboard whose widgets are AI-generated from source sheets	Read / Write	create_dashboard reads sheet data directly from the supplied asset IDs and rolls back automatically on failure. Needs CREATE_SIGHTS and a workspace destination
Discovery	Searching across every accessible Smartsheet asset by name or content, scoped to sheet, report, dashboard, folder and workspace names, cell data, comments, attachments, summary fields and templates	Read	The cellData, comments and attachments scopes search inside content rather than names, so this reaches real data; it returns up to 100 results with no pagination
Solutions	An AI agent that builds end-to-end solutions from a natural language description, creating any combination of workspaces, sheets, dashboards, workflows, reports, forms, portfolios, projects and blueprints in one call	Read / Write	The broadest tool on the server: one call can create many assets at once, takes 1 to 2 minutes or longer, and makes best-effort interpretations of ambiguous requirements. It holds five access scopes at once
Favorites	Listing and checking the current user's favorites, and adding or removing them	Read / Write	Low risk; affects only the current user
Portfolios & Projects	Portfolios, project templates and projects, and the portfolio project list sheet	Read / Write	Deleting a portfolio or template removes its workspaces; require approval
Scenario Planning	Scenario plans and scenarios, scenario sheets and rows, diffs, capacity and workload, and applying a scenario	Read / Write	apply_scenario changes the source sheets; require approval
Plans & Licensing	Plan IDs, per-user seat usage and license inventory	Read	Admin data; limit to plan administrators

Connect Smartsheet MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect Smartsheet
Find Smartsheet in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow Smartsheet MCP or block it. You allow the exact tools each agent can call, one by one.

Smartsheet MCP	83 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 83 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is Smartsheet MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	Smartsheet MCP83 tools from Smartsheet's published reference, which presents the key tools rather than a guaranteed complete list. Requires a Business, Enterprise or Advanced Work Management plan.
Coding assistants	Connection & credentialsper user	Scenario Planning13 tools
Internal copilots	Tool access policiesdeny by default	Portfolios & Projects11 tools
Workflow automations	Request & response inspectioninline	Charts8 tools
Compatible MCP clients	Activity monitoringalerts and logs	All other tools51 tools