daemon.cjs · part 222

60 text occurrences from desktop/Cursor.app/Contents/Resources/app/extensions/cursor-agent-host/dist/agent-host-daemon/dist/bin/daemon.cjs, part 222. Every entry preserves the shipped literal and its saved verdict or selection reason.

File contents and all parts · All files

<transcript_location>
${xmlContent}
</transcript_location>

### Transcript location:
  This is the full JSONL transcript of your past conversation with the user (pre- and post-summary): ${transcriptPath}

  If anything about the task or current state is unclear (missing context, ambiguous requirements, uncertain decisions, exact wording, IDs/paths, errors/logs), you should consult this transcript.

  How to use it:
  - Search first for relevant keywords (task name, filenames, IDs, errors, tool names).
  - Then read a small window around the matching lines to reconstruct intent and state.
  - Avoid reading linearly end-to-end; the file can be very large and some single lines can be huge.
  - Files contain one structured json event per line including user/assistant messages. Currently tool calls and results are excluded.
  

### Transcript location:
  - This folder contains full transcripts of past conversations with the user (pre- and post-summary): ${agentTranscriptsFolder}
  - Each conversation is stored as a <convoId>/<convoId>.jsonl.

use strict

<todo_update>
${todoContent}
</todo_update>

use strict

The below plan was previously created in this session. If you are still in plan mode, continue to iterate on the plan with the user given the rest of your context on the current conversation. Otherwise, if there are remaining relevant todos, you should continue to implement them according to the plan.

10. Current plan mode progress:
   ${instruction}

<file_contents path="${planFilePath(enrichments.currentPlan)}" isFullFile="true">
${content}
</file_contents>

<current_plan>
${content}
</current_plan>

NOTE: This is an automation run. The original trigger info that started this session:
${enrichments.automationTriggerContext}

12. Automation trigger info (this is the original trigger that started this agent):

${enrichments.automationTriggerContext}

NOTE: There was an active todo list in the conversation. Here is the latest update before summarization:
${todoUpdateTag(enrichments.todoContent)}

request contains text fields that are too large

not valid json

invalid json

invalid argument

User API Key Rate limit exceeded

use strict

No assistant response received

Context summarization on ${details.summarizerModelId} failed (${details.errorKind}); the previous context was kept

request contained too many images or documents

too many images or documents error

too many images and documents

too much media

use strict

[summarization-pipeline] preservedTailMessages shape invariant violated

use strict

use strict

use strict

[summarization-handler] Error getting extended usage

[summarization-handler] Error getting usage

[summarization-handler] Error getting provider metadata

[summarization-handler] Error getting invocation id

[summarization-handler] Error getting response

[summarization-handler] Summarization response contained more than one assistant message; using only the last to avoid duplicated summaries

Summarization returned empty text

${prefix} ${truncated}
[... truncated, ${serializedValue.length} total chars]

${prefix} ${serializedValue}

[Thinking] ${c.text}

[Tool call] ${c.toolName}

[Tool call] ${c.toolName} ${argsStr}

[Tool result] ${c.toolName}

[Tool result] ${c.toolName} ${resultStr}

<summarization_request>${MORE_PROMPT}${options2?.requestShorterOutput ? SHORTER_OUTPUT_RETRY_PROMPT : ""}</summarization_request>

${m2.role}: ${turnMessageToString(m2, true)}

[summarization-handler] Reducing prompt budget due to byte limit

[summarization-handler] Would truncate summarization prompt

[summarization-handler] failed to build summarization prompt

Cannot fit any summarization message within prompt char budget

[summarization-handler] Suspiciously short joined result length for summarization

[summarization-handler] Final prompt exceeded byte budget

[summarization-handler] Summarization prompt exceeded char budget; truncated

[summarization-handler] Attempt started

[summarization-handler] Stripped non-printable characters from prompt on final input-limit retry

[summarization-handler] Expected prompt truncation after input limit reduction but prompt fit within reduced budget

[summarization-handler] Attempt succeeded

[summarization-handler] prompt guard did not reduce inputs enough

[summarization-handler] Error summarizing messages

[summarization-handler] Retrying with shorter-output instruction

[summarization-handler] Retrying with reduced input budget