daemon.cjs · part 350

60 text occurrences from desktop/Cursor.app/Contents/Resources/app/extensions/cursor-agent-host/dist/agent-host-daemon/dist/bin/daemon.cjs, part 350. Every entry preserves the shipped literal and its saved verdict or selection reason.

File contents and all parts · All files

Discover and inspect tools available through dynamic namespaces, e.g. MCP servers.

1. {"namespace":"<id>"}: returns full input schemas and full descriptions for every tool in that namespace.

2. {"namespace":"<id>","toolName":"<name>"}: returns the full schema and full description for one tool.

3. {"pattern":"<regex>"}: searches namespace and tool names across all namespaces using RE2 syntax.

4. {"namespace":"<id>","pattern":"<regex>"}: searches tool names within that namespace.

5. No arguments: returns the full catalog. Prefer a namespace or pattern when possible.

Pattern-search and catalog results shorten long descriptions to 200 characters, ending with "${TRUNCATED_DESCRIPTION_SUFFIX}". Namespace and single-tool lookups always return the complete description, so fetch the tool directly when you need the full text.

The response includes namespaceStatus for MCP-backed namespaces; do not treat namespaces in ${unusableStatusList} states as usable.

Always call this tool to discover a tool's schema before calling it with ${callMcpToolName}.

Discover and inspect MCP tools. There are 5 ways to call this tool. Prefer fetching by server or pattern over listing the full catalog.

1. {"server":"<id>"}: returns full input schemas and full descriptions for every tool on that server. Preferred when you know the server.

2. {"server":"<id>","toolName":"<name>"}: returns the full schema and full description for one tool.

3. {"pattern":"<regex>"}: searches tool and server names across all servers using RE2 syntax.

4. {"server":"<id>","pattern":"<regex>"}: searches tool names on that server using RE2 syntax.

5. No arguments: returns a catalog of all servers with tool names and short descriptions. Use only as a last resort.

Pattern-search and catalog results shorten long descriptions to 200 characters, ending with "${TRUNCATED_DESCRIPTION_SUFFIX}". Server and single-tool lookups always return the complete description, so fetch the tool directly when you need the full text.

The response includes each server's serverStatus; do not treat servers in ${unusableStatusList} states as usable.

Always call this tool to discover a tool's schema before calling it with ${callMcpToolName}.

Invalid arguments

use strict

Object expected.

Symbol.asyncDispose is not defined.

Symbol.dispose is not defined.

Object not disposable.

An error was suppressed during disposal.

List MCP resources rejected

List MCP resources executor returned result without case

The MCP resources executor returned an empty result. This usually means the client is outdated or incompatible; update the client and retry.

Unexpected result case: ${_exhaustive}

Unknown error

Error listing MCP resources: ${execResult.result.value.error}

List MCP resources rejected: ${execResult.result.value.reason}

Unknown error

No MCP resources found.

• ${resource.name || resource.uri}

 - ${resource.description}

 (${resource.server})

 [${resource.mimeType}]

  URI: ${resource.uri}

Found ${resources.length} MCP resource(s):
${resourceList}

List available resources from configured MCP servers. Each returned resource will include all standard MCP resource fields plus a 'server' field indicating which server the resource belongs to. MCP resources are _not_ the same as tools, so don't call this function to discover MCP tools.

Optional server identifier to filter resources by. If not provided, resources from all servers will be returned.

[Reference error: deprecated syntax, use $PROMPT_<user_message_id>]

[Reference error: no prompt found for user message id ${promptId}]

[Reference error: no response found for subagent ${subagentId}]

[Reference error: invalid reference syntax]

use strict

Optional array of file paths to image/video files to attach to your prompt to the ${target}. 

Failed to create agent

Failed to send prompt to agent

${prefix}: ${result.value.error}

Failed to send prompt to agent: ${RUNNING_SUBAGENT_FOLLOWUP_ERROR}

 ${result.value.resultSuffix}

Created agent with agent_id: ${taskId}. It is running in the background.${suffix}

Sent the prompt to agent_id: ${taskId}. It is running in the background.${suffix}

Agent response was empty.

Unknown agent response.

agent_id: 

agent_id: 

Meta parent agents require client-side subagent execution.