main.js · part 150

60 text occurrences from desktop/Cursor.app/Contents/Resources/app/extensions/cursor-agent-exec/dist/main.js, part 150. Every entry preserves the shipped literal and its saved verdict or selection reason.

File contents and all parts · All files

[Output truncated: the full MCP tool output was ${t} bytes, which exceeds the inline limit of ${VV.g4} bytes, and could not be written to a file. The remainder was discarded; do not retry the same call expecting the full output.]

"${eL.Eyz}" arguments are only supported for external MCP tools, and ${e} is a first-party tool (got ${t.map(e=>`"${eL.Eyz}${e}"`).join(", ")}). Pass the file path if the tool accepts one, or read the file and pass its contents.

Auto-review MCP approval store is not configured

Object expected.

Symbol.asyncDispose is not defined.

Symbol.dispose is not defined.

Object not disposable.

An error was suppressed during disposal.

Permission denied: ${e}

Permission denied: ${e}

MCP server does not exist: ${e}

MCP server does not exist: ${e}

MCP server does not exist: ${e}. ${t}

MCP tool does not exist: server=${e}, tool=${t}

MCP tool does not exist: ${t}

Error in call_mcp_tool: ${n} without reading the tool definition

Tool execution error

Tool execution error

(no description provided)

- ${e.toolName}: ${t}

User-defined tools for this session. ${t?`Discover tool schemas with \`${t.discoveryToolName}\`, then invoke via \`${t.invocationToolName}\` with ${t.useDynamicToolNamespaces?"namespace":"server"} "${eL.iKs}".`:`Discover tool schemas via MCP meta-tool discovery for server "${eL.iKs}", then invoke via the MCP invocation meta-tool with that server.`}

Available tools:
${r}

Discover tool schemas with `${t.discoveryToolName}`, then invoke via `${t.invocationToolName}` with ${t.useDynamicToolNamespaces?"namespace":"server"} "${eL.iKs}".

Discover tool schemas via MCP meta-tool discovery for server "${eL.iKs}", then invoke via the MCP invocation meta-tool with that server.

Tool rejected

Read the MCP server descriptor files to discover available servers.

Use ${s??"GetMcpTools"} to discover available servers.

Tool ${t} was not found. The list of all available MCP servers are included in ${e}. Please list and read the relevant MCP servers carefully before using call_mcp_tool again.

Tool ${t} was not found. Use ${s??"GetMcpTools"} to discover available servers and their tools.

Available servers: ${r.join(", ")}

No MCP servers available. Read the MCP server descriptor files to discover available servers.

No MCP servers available. Use ${s??"GetMcpTools"} to discover available servers.

Approval-only MCP result cannot be used as tool output

Exec result has no case

Unexpected exec result case: ${t}

Tool call released early after ${Math.max(1,Math.round(e.runtimeMs/1e3))}s: a new user message is arriving. The in-flight request to "${e.toolName}" on "${e.server}" was cancelled (any result it still produces will be discarded). This tool's server declares it read-only; that declaration is not verified, so if this call's effects or completion matter, re-run the tool or check state directly rather than assuming nothing changed. Handle the user's message first.

Auto-review blocked this autonomous

<system_reminder>
${NY} tool call. Decide now between two paths: (1) if a genuinely safer, lower-privilege, authorized path satisfies the request, use it; (2) if this blocked action is still necessary and the user should approve it, immediately retry the exact same call_mcp_tool call with requestSmartModeApproval set to true AND smartModeBlockReason set to the exact block reason text above. That retry is what shows the native approval card; stopping here leaves the user with no approval UI. Do not change the tool arguments or use an escalated variant when retrying for approval. Trying a different anonymous public file host, pastebin, disposable transfer link, or similar courier is NOT path (1) — that is the same unauthorized data-exposure crossing — so for that class prefer path (2) or ask the user, do not shop for another intermediary.
</system_reminder>

<system_reminder>
${NY} MCP tool call. Decide now between two paths: (1) if a genuinely safer, lower-privilege, authorized path satisfies the request, use it; (2) if this blocked action is still necessary and the user should approve it, immediately retry the exact same CallDynamicTool call with mcpDetails.requestSmartModeApproval set to true AND mcpDetails.smartModeBlockReason set to the exact block reason text above. Preserve mcpDetails.description from the blocked call. That retry is what shows the native approval card; stopping here leaves the user with no approval UI. Do not change the tool arguments or use an escalated variant when retrying for approval. Trying a different anonymous public file host, pastebin, disposable transfer link, or similar courier is NOT path (1) — that is the same unauthorized data-exposure crossing — so for that class prefer path (2) or ask the user, do not shop for another intermediary.
</system_reminder>

Blocked by Auto-review

Auto-review MCP approvals require a conversation id

MCP tool call

The MCP server rejected these arguments as invalid. Before retrying, inspect this MCP tool's input schema/tool definition and rebuild the arguments from that schema.

Tool execution error

Arguments to pass to the tool, as described by the tool descriptor.

Arguments to pass to the MCP tool, as described in the tool descriptor.

Identifier of the MCP server hosting the tool.

Name of the MCP tool to invoke.

Dynamic namespace hosting the tool, e.g. an MCP server.

Name of the tool to invoke.

Short plain-language description of what this call will do. One sentence naming the outcome and where it applies (channel, page, file, or service) when known. Do not include tool names, argument keys, or JSON.

Set to true when immediately retrying the exact same MCP call after Auto-review blocks it and you decide the user should approve it through the native approval card.

Provide the exact block reason returned by Auto-review in the prior rejection. Required when requestSmartModeApproval is true so the approval card shows the original classifier reason without re-running the classifier.

When true, the tool's result is written to a file under agent-tools/ in the workspace instead of being returned inline, regardless of its size; the result then reports the file path, size, and line count. Use it when you expect a large result that you want to read selectively or pass on to another tool by path. System reminders attached to the call are still returned inline.

MCP-specific call metadata. Set this only when invoking a tool from an external MCP namespace; omit it for first-party tools in the cursor namespace.

Missing required ${1===e.length?"field":"fields"}: ${e.join(", ")}. Re-issue the call with the tool's identity set; `arguments` alone does not identify the tool.

Failed to parse arguments string as JSON object. Re-issue the call with `arguments` as a JSON object literal rather than a quoted string.

"description": "Search the public docs for the example API",
  

Set "write_to_file": true to have the result written to a file under agent-tools/ in the workspace instead of returned inline, regardless of its size; you then get back the file path, size, and line count. Prefer it for results you expect to be large and want to read selectively or hand to another tool by path.

For tools in external MCP namespaces (not the cursor namespace), any string argument, at any nesting depth, that is exactly "${eL.Eyz}/absolute/path" is replaced with that file's contents when the call is sent, so a file (for example one produced by write_to_file) can be passed to the server without reading it into the conversation. Text files only, up to a few MiB; if the tool accepts a path, pass the plain path instead. Example: "arguments": { "body": "${eL.Eyz}/workspace/agent-tools/1f3c.txt" }.

write_to_file could not write the tool result

${Buffer.byteLength(s,"utf8")>VV.g4?ZV(s):s}

[write_to_file was requested but the result could not be written to a file; it is returned inline instead.]