main.js · part 44

60 text occurrences from desktop/Cursor.app/Contents/Resources/app/extensions/cursor-agent-host/dist/main.js, part 44. Every entry preserves the shipped literal and its saved verdict or selection reason.

File contents and all parts · All files

${e} [git: ${s.join(" ")}]

--- git stderr ---
${i=(0,a.ix)(n.stderr),i.length<=u?i:`${i.slice(0,u)}… [truncated to 2048 chars]`}

${i.slice(0,u)}… [truncated to 2048 chars]

Duration of controlled exec operations in milliseconds

Count of successful controlled exec operations

Count of failed controlled exec operations

Computer use is not enabled for this environment, so the Computer tool and the computerUse subagent cannot run here. Do not retry computer actions; verify with non-GUI methods (automated tests, shell commands) instead. (No computer-use handler is registered on this exec daemon.)

Screen recording is not enabled for this environment, so recordings cannot be captured here. Do not retry; skip the recording and present other evidence instead. (No record-screen handler is registered on this exec daemon.)

Unknown error

No handler found for server message of type ${a}

Object expected.

Symbol.asyncDispose is not defined.

Symbol.dispose is not defined.

Object not disposable.

An error was suppressed during disposal.

Hook additional_context carriers drained into the caller collector

No exec result

No result value

Ignoring exec stream shutdown during detached drain

Cursor Slack Tools

Cursor Automation Tools

This MCP server requires authentication before its tools can be used.

This MCP server failed during live tool discovery. Its tools are unavailable until the connection is fixed.

This MCP server is still loading; tools may not be available yet.

Outcome of late-binding $file:/abs/path MCP argument expansion at the executor boundary

Total bytes substituted into one MCP tool call by $file:/abs/path arguments

Number of distinct $file:/abs/path arguments expanded in one MCP tool call

Could not resolve the symbolic argument "${Gt}${t}": ${r} A string argument of the form "${Gt}/absolute/path" is replaced with that file's contents before the tool runs; pass the literal text instead if you did not intend that.

the file is not a text file.

the file is too large to inline (${Number(e.fileSize)||n} bytes; at most ${s} bytes per file may be substituted). Pass the path to a tool that accepts one, or read a range of it.

the file does not exist.

reading it was not permitted.

${o.result.value.reason||"the read was rejected"}. Only files the Read tool can read without approval can be substituted; read it with Read instead.

the read was rejected

the file cannot be read as text.

the read returned no result.

at most ${Wt} symbolic arguments may be used in one call (${t.length} given). Split the call or pass paths to tools that accept them.

this environment cannot read files on behalf of a tool call.

together the ${t.length} symbolic arguments exceed the ${o} bytes that may be substituted per call. Pass paths to tools that accept them instead.

the file changed after the call was reviewed (${r?.sizeBytes??0} bytes then, ${t.bytes} bytes now), so its new contents were not sent. Re-run the call to review the current contents.

${t}

The request carried ${r.bytes} bytes of file contents substituted from ${r.paths} "${Gt}/abs/path" argument(s), and the server appears to have rejected the request body as too large. Do not retry the same call: pass a smaller file or a range of it, or use a tool that accepts a path or upload instead.

"${Gt}${e.path}" was replaced with the contents of ${e.path} (${e.bytes} bytes) before the call.

This operation is not allowed in readonly mode. The subagent was launched with readonly: true, which restricts write operations.

This operation is not allowed in readonly mode. The subagent was launched with readonly: true, which restricts shell, write, delete, and MCP operations.

${e} Tool: ${r.name}

Sub-agent is currently running. You may send the follow-up message when it has completed.

Too many local subagents are already running (limit: ${e.maxRunning}). Wait for running subagents to finish (or collect background subagent results) before starting another one.

If you intended to interrupt this agent, you may retry with `interrupt` set to true

Subagent was aborted by the user

Plan written to ${e.tool.value.result?.planUri?.trim()||"unknown plan file"}:
${t}

unknown plan file

No output

Unhandled idle-state event: ${String(t)}

Unhandled running-state event: ${String(t)}

Unhandled runningWithQueuedResume-state event: ${String(t)}

Unhandled lifecycle manager effect: ${String(t)}

--- Cursor Command: ${e} ---

<cursor_commands>
${r.map(e=>`${vn(e.name)}\n${e.content}\n--- End Command ---`).join("\n")}
</cursor_commands>

${e}

${vn(e.name)}
${e.content}
--- End Command ---

use strict