test: update unit test snapshots

fix: resolve duplicate step numbering in replace_in_file instructions
Fixes the duplicate step 5 issue when NOTEBOOK_INSTRUCTIONS is concatenated with BASE_DIFF_INSTRUCTIONS for .ipynb files. Changes notebook instructions to step 6 to avoid ambiguity.
2026-09-02 07:42:19 +08:00 · 2026-03-26 17:35:34 -07:00 · 2026-03-26 12:18:13 -07:00 · 2026-03-26 12:17:19 -07:00 · 2026-03-06 12:20:24 -08:00 · 2026-03-06 12:20:24 -08:00
29 changed files with 358 additions and 50 deletions
@@ -54,6 +54,8 @@ export const toolParamNames = [
 	"prompt_3",
 	"prompt_4",
 	"prompt_5",
+	"start_line",
+	"end_line",
 ] as const

 export type ToolParamName = (typeof toolParamNames)[number]
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -36,12 +36,16 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 </read_file>

 ## write_to_file
@@ -83,6 +87,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 Usage:
 <replace_in_file>
 <path>File path here</path>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -36,12 +36,16 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 </read_file>

 ## write_to_file
@@ -83,6 +87,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 Usage:
 <replace_in_file>
 <path>File path here</path>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -59,7 +59,7 @@
    "type": "function",
    "function": {
      "name": "read_file",
-      "description": "Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
+      "description": "Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
      "strict": false,
      "parameters": {
        "type": "object",
@@ -68,6 +68,14 @@
            "type": "string",
            "description": "The path of the file to read (relative to the current working directory {{CWD}}){{MULTI_ROOT_HINT}}"
          },
+          "start_line": {
+            "type": "integer",
+            "description": "The 1-based line number to start reading from (inclusive). Defaults to 1."
+          },
+          "end_line": {
+            "type": "integer",
+            "description": "The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files."
+          },
          "task_progress": {
            "type": "string",
            "description": "A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)"
@@ -125,7 +133,7 @@
          },
          "diff": {
            "type": "string",
-            "description": "One or more SEARCH/REPLACE blocks following this exact format:\n  ```\n  ------- SEARCH\n  [exact content to find]\n  =======\n  [new content to replace with]\n  +++++++ REPLACE\n  ```\n  Critical rules:\n  1. SEARCH content must match the associated file section to find EXACTLY:\n     * Match character-for-character including whitespace, indentation, line endings\n     * Include all comments, docstrings, etc.\n  2. SEARCH/REPLACE blocks will ONLY replace the first match occurrence.\n     * Including multiple unique SEARCH/REPLACE blocks if you need to make multiple changes.\n     * Include *just* enough lines in each SEARCH section to uniquely match each set of lines that need to change.\n     * When using multiple SEARCH/REPLACE blocks, list them in the order they appear in the file.\n  3. Keep SEARCH/REPLACE blocks concise:\n     * Break large SEARCH/REPLACE blocks into a series of smaller blocks that each change a small portion of the file.\n     * Include just the changing lines, and a few surrounding lines if needed for uniqueness.\n     * Do not include long runs of unchanging lines in SEARCH/REPLACE blocks.\n     * Each line must be complete. Never truncate lines mid-way through as this can cause matching failures.\n  4. Special operations:\n     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)\n     * To delete code: Use empty REPLACE section"
+            "description": "One or more SEARCH/REPLACE blocks following this exact format:\n  ```\n  ------- SEARCH\n  [exact content to find]\n  =======\n  [new content to replace with]\n  +++++++ REPLACE\n  ```\n  Critical rules:\n  1. SEARCH content must match the associated file section to find EXACTLY:\n     * Match character-for-character including whitespace, indentation, line endings\n     * Include all comments, docstrings, etc.\n  2. SEARCH/REPLACE blocks will ONLY replace the first match occurrence.\n     * Including multiple unique SEARCH/REPLACE blocks if you need to make multiple changes.\n     * Include *just* enough lines in each SEARCH section to uniquely match each set of lines that need to change.\n     * When using multiple SEARCH/REPLACE blocks, list them in the order they appear in the file.\n  3. Keep SEARCH/REPLACE blocks concise:\n     * Break large SEARCH/REPLACE blocks into a series of smaller blocks that each change a small portion of the file.\n     * Include just the changing lines, and a few surrounding lines if needed for uniqueness.\n     * Do not include long runs of unchanging lines in SEARCH/REPLACE blocks.\n     * Each line must be complete. Never truncate lines mid-way through as this can cause matching failures.\n  4. Special operations:\n     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)\n     * To delete code: Use empty REPLACE section\n  5. If your source context came from read_file and includes line labels (for example, \"42 | const x = 1\"), do NOT include the \"42 | \" prefix in SEARCH or REPLACE content. Match only the raw file text."
          },
          "task_progress": {
            "type": "string",
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -36,12 +36,16 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 </read_file>

 ## write_to_file
@@ -83,6 +87,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 Usage:
 <replace_in_file>
 <path>File path here</path>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -36,12 +36,16 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 </read_file>

 ## write_to_file
@@ -83,6 +87,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 Usage:
 <replace_in_file>
 <path>File path here</path>
@@ -39,13 +39,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -90,6 +94,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -29,7 +29,7 @@
    "type": "function",
    "function": {
      "name": "read_file",
-      "description": "Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
+      "description": "Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
      "strict": false,
      "parameters": {
        "type": "object",
@@ -38,6 +38,14 @@
            "type": "string",
            "description": "The path of the file to read (relative to the current working directory {{CWD}}){{MULTI_ROOT_HINT}}"
          },
+          "start_line": {
+            "type": "integer",
+            "description": "The 1-based line number to start reading from (inclusive). Defaults to 1."
+          },
+          "end_line": {
+            "type": "integer",
+            "description": "The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files."
+          },
          "task_progress": {
            "type": "string",
            "description": "A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)"
@@ -29,7 +29,7 @@
    "type": "function",
    "function": {
      "name": "read_file",
-      "description": "Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
+      "description": "Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
      "strict": false,
      "parameters": {
        "type": "object",
@@ -38,6 +38,14 @@
            "type": "string",
            "description": "The path of the file to read (relative to the current working directory {{CWD}}){{MULTI_ROOT_HINT}}"
          },
+          "start_line": {
+            "type": "integer",
+            "description": "The 1-based line number to start reading from (inclusive). Defaults to 1."
+          },
+          "end_line": {
+            "type": "integer",
+            "description": "The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files."
+          },
          "task_progress": {
            "type": "string",
            "description": "A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)"
@@ -48,13 +48,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -99,6 +103,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -48,13 +48,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -99,6 +103,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -45,12 +45,16 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 </read_file>

 ## write_to_file
@@ -92,6 +96,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 Usage:
 <replace_in_file>
 <path>File path here</path>
@@ -48,13 +48,17 @@ Usage:
 </execute_command>

 ## read_file
-Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
+Description: Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.
 Parameters:
 - path: (required) The path of the file to read (relative to the current working directory /test/project)
+- start_line: (optional) The 1-based line number to start reading from (inclusive). Defaults to 1.
+- end_line: (optional) The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <read_file>
 <path>File path here</path>
+<start_line>1</start_line>
+<end_line>1000</end_line>
 <task_progress>Checklist here (optional)</task_progress>
 </read_file>

@@ -99,6 +103,7 @@ Parameters:
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.
 - task_progress: (optional) A checklist showing task progress after this tool use is completed. The task_progress parameter must be included as a separate parameter inside of the parent tool call, it must be separate from other parameters such as content, arguments, etc. (See 'UPDATING TASK PROGRESS' section for more details)
 Usage:
 <replace_in_file>
@@ -20,13 +20,19 @@
  },
  {
    "name": "read_file",
-    "description": "Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
+    "description": "Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
    "parameters": {
      "type": "OBJECT",
      "properties": {
        "path": {
          "type": "STRING"
        },
+        "start_line": {
+          "type": "NUMBER"
+        },
+        "end_line": {
+          "type": "NUMBER"
+        },
        "task_progress": {
          "type": "STRING"
        }
@@ -5,38 +5,48 @@ import { TASK_PROGRESS_PARAMETER } from "../types"

 const id = ClineDefaultTool.FILE_READ

+const READ_FILE_DESCRIPTION =
+	"Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Returned text lines are prefixed with line labels (e.g. `1 |`, `2 |`). These labels are metadata, not part of the file content. For large files, output is automatically limited to 1000 lines. Use start_line and end_line to read specific sections. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files."
+
+const READ_FILE_PARAMETERS: ClineToolSpec["parameters"] = [
+	{
+		name: "path",
+		required: true,
+		instruction: `The path of the file to read (relative to the current working directory {{CWD}}){{MULTI_ROOT_HINT}}`,
+		usage: "File path here",
+	},
+	{
+		name: "start_line",
+		required: false,
+		type: "integer",
+		instruction: "The 1-based line number to start reading from (inclusive). Defaults to 1.",
+		usage: "1",
+	},
+	{
+		name: "end_line",
+		required: false,
+		type: "integer",
+		instruction:
+			"The 1-based line number to stop reading at (inclusive). Defaults to start_line + 1000. Use with start_line to read specific sections of large files.",
+		usage: "1000",
+	},
+	TASK_PROGRESS_PARAMETER,
+]
+
 const generic: ClineToolSpec = {
 	variant: ModelFamily.GENERIC,
 	id,
 	name: "read_file",
-	description:
-		"Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
-	parameters: [
-		{
-			name: "path",
-			required: true,
-			instruction: `The path of the file to read (relative to the current working directory {{CWD}}){{MULTI_ROOT_HINT}}`,
-			usage: "File path here",
-		},
-		TASK_PROGRESS_PARAMETER,
-	],
+	description: READ_FILE_DESCRIPTION,
+	parameters: READ_FILE_PARAMETERS,
 }

 const NATIVE_GPT_5: ClineToolSpec = {
 	variant: ModelFamily.NATIVE_GPT_5,
 	id,
 	name: "read_file",
-	description:
-		"Request to read the contents of a file at the specified path. Use this when you need to examine the contents of an existing file you do not know the contents of, for example to analyze code, review text files, or extract information from configuration files. Automatically extracts raw text from PDF and DOCX files. May not be suitable for other types of binary files, as it returns the raw content as a string. Do NOT use this tool to list the contents of a directory. Only use this tool on files.",
-	parameters: [
-		{
-			name: "path",
-			required: true,
-			instruction: `The path of the file to read (relative to the current working directory {{CWD}}){{MULTI_ROOT_HINT}}`,
-			usage: "File path here",
-		},
-		TASK_PROGRESS_PARAMETER,
-	],
+	description: READ_FILE_DESCRIPTION,
+	parameters: READ_FILE_PARAMETERS,
 }

 const NATIVE_NEXT_GEN: ClineToolSpec = {
@@ -36,10 +36,11 @@ const BASE_DIFF_INSTRUCTIONS = `One or more SEARCH/REPLACE blocks following this
     * Each line must be complete. Never truncate lines mid-way through as this can cause matching failures.
  4. Special operations:
     * To move code: Use two SEARCH/REPLACE blocks (one to delete from original + one to insert at new location)
-     * To delete code: Use empty REPLACE section`
+     * To delete code: Use empty REPLACE section
+  5. If your source context came from read_file and includes line labels (for example, "42 | const x = 1"), do NOT include the "42 | " prefix in SEARCH or REPLACE content. Match only the raw file text.`

 const NOTEBOOK_INSTRUCTIONS = `
-  5. For Jupyter Notebook (.ipynb) files:
+  6. For Jupyter Notebook (.ipynb) files:
     * Match the exact JSON structure including quotes, commas, and \\n characters
     * Each line in "source" array (except last) must end with "\\n"
     * Each source line is a separate JSON string in the array
@@ -15,6 +15,51 @@ import type { TaskConfig } from "../types/TaskConfig"
 import type { StronglyTypedUIHelpers } from "../types/UIHelpers"
 import { ToolResultUtils } from "../utils/ToolResultUtils"

+export const DEFAULT_MAX_LINES = 1000
+const FILE_TRUNCATED_MARKER = "\n\n---\n\n[FILE TRUNCATED:"
+
+/**
+ * Slice file content to the requested line range, add L-prefixed line labels,
+ * and append a continuation hint when the file has more lines to read.
+ */
+export function formatFileContentWithLineNumbers(content: string, startLine?: number, endLine?: number): string {
+	if (!content) {
+		return content
+	}
+
+	// Separate any byte-truncation notice appended by content-limits.ts
+	let body = content
+	let truncationSuffix = ""
+	const truncationIndex = content.indexOf(FILE_TRUNCATED_MARKER)
+	if (truncationIndex !== -1) {
+		body = content.slice(0, truncationIndex)
+		truncationSuffix = content.slice(truncationIndex)
+	}
+
+	const lines = body.split(/\r?\n/)
+	if (body.endsWith("\n") && lines.length > 0) {
+		lines.pop()
+	}
+	const totalLines = lines.length
+
+	const start = Math.max(1, startLine ?? 1)
+	const end = Math.min(totalLines, endLine ?? start + DEFAULT_MAX_LINES - 1)
+
+	const slice = lines.slice(start - 1, end)
+	const labeled = slice.map((line, i) => `${start + i} | ${line}`).join("\n")
+
+	let suffix = truncationSuffix
+	if (!truncationSuffix) {
+		if (end < totalLines) {
+			suffix = `\n\n(Showing lines ${start}-${end} of ${totalLines} total. Use start_line=${end + 1} to continue reading.)`
+		} else {
+			suffix = `\n\n(File has ${totalLines} lines total.)`
+		}
+	}
+
+	return labeled + suffix
+}
+
 export class ReadFileToolHandler implements IFullyManagedTool {
 	readonly name = ClineDefaultTool.FILE_READ

@@ -179,8 +224,17 @@ export class ReadFileToolHandler implements IFullyManagedTool {
 		// Handle image blocks separately - they need to be pushed to userMessageContent
 		if (fileContent.imageBlock) {
 			config.taskState.userMessageContent.push(fileContent.imageBlock)
+			return fileContent.text
 		}

-		return fileContent.text
+		// Parse and validate line parameters with proper NaN guards
+		const startLine = block.params.start_line ? Number.parseInt(block.params.start_line, 10) : undefined
+		const endLine = block.params.end_line ? Number.parseInt(block.params.end_line, 10) : undefined
+		
+		// Validate parsed numbers to prevent NaN propagation
+		const safeStartLine = startLine !== undefined && !Number.isNaN(startLine) ? startLine : undefined
+		const safeEndLine = endLine !== undefined && !Number.isNaN(endLine) ? endLine : undefined
+
+		return formatFileContentWithLineNumbers(fileContent.text, safeStartLine, safeEndLine)
 	}
 }
@@ -0,0 +1,111 @@
+import assert from "node:assert/strict"
+import { describe, it } from "mocha"
+import { DEFAULT_MAX_LINES, formatFileContentWithLineNumbers } from "../ReadFileToolHandler"
+
+describe("formatFileContentWithLineNumbers", () => {
+	describe("line labels", () => {
+		it("adds 1-indexed line prefixes", () => {
+			const result = formatFileContentWithLineNumbers("alpha\nbeta")
+			assert.ok(result.startsWith("1 | alpha\n2 | beta"))
+		})
+
+		it("does not add an extra numbered line for trailing newline", () => {
+			const result = formatFileContentWithLineNumbers("alpha\nbeta\n")
+			assert.ok(result.startsWith("1 | alpha\n2 | beta"))
+			assert.ok(!result.includes("3 |"))
+		})
+
+		it("returns empty content unchanged", () => {
+			const result = formatFileContentWithLineNumbers("")
+			assert.equal(result, "")
+		})
+	})
+
+	describe("chunked reads", () => {
+		const tenLines = Array.from({ length: 10 }, (_, i) => `line${i + 1}`).join("\n")
+
+		it("defaults to reading from line 1", () => {
+			const result = formatFileContentWithLineNumbers(tenLines)
+			assert.ok(result.startsWith("1 | line1\n"))
+		})
+
+		it("respects start_line parameter", () => {
+			const result = formatFileContentWithLineNumbers(tenLines, 5)
+			assert.ok(result.startsWith("5 | line5\n"))
+			assert.ok(!result.includes("4 |"))
+		})
+
+		it("respects start_line and end_line parameters", () => {
+			const result = formatFileContentWithLineNumbers(tenLines, 3, 5)
+			assert.ok(result.startsWith("3 | line3\n"))
+			assert.ok(result.includes("4 | line4\n"))
+			assert.ok(result.includes("5 | line5"))
+			assert.ok(!result.includes("2 |"))
+			assert.ok(!result.includes("6 |"))
+		})
+
+		it("clamps start_line to 1 if below", () => {
+			const result = formatFileContentWithLineNumbers(tenLines, -5, 3)
+			assert.ok(result.startsWith("1 | line1\n"))
+		})
+
+		it("clamps end_line to total lines if beyond", () => {
+			const result = formatFileContentWithLineNumbers(tenLines, 8, 999)
+			assert.ok(result.includes("10 | line10"))
+			assert.ok(!result.includes("11 |"))
+		})
+	})
+
+	describe("continuation hints", () => {
+		const tenLines = Array.from({ length: 10 }, (_, i) => `line${i + 1}`).join("\n")
+
+		it("shows continuation hint when more lines remain", () => {
+			const result = formatFileContentWithLineNumbers(tenLines, 1, 5)
+			assert.ok(result.includes("Showing lines 1-5 of 10 total"))
+			assert.ok(result.includes("start_line=6"))
+		})
+
+		it("shows total-lines footer when entire file is returned", () => {
+			const result = formatFileContentWithLineNumbers(tenLines, 1, 10)
+			assert.ok(result.includes("File has 10 lines total"))
+		})
+
+		it("shows total-lines footer when file fits within default limit", () => {
+			const result = formatFileContentWithLineNumbers(tenLines)
+			assert.ok(result.includes("File has 10 lines total"))
+		})
+	})
+
+	describe("default max lines", () => {
+		const bigContent = Array.from({ length: DEFAULT_MAX_LINES + 500 }, (_, i) => `row${i + 1}`).join("\n")
+
+		it("limits output to DEFAULT_MAX_LINES when no end_line given", () => {
+			const result = formatFileContentWithLineNumbers(bigContent)
+			assert.ok(result.includes(`1 | row1`))
+			assert.ok(result.includes(`${DEFAULT_MAX_LINES} | row${DEFAULT_MAX_LINES}`))
+			assert.ok(!result.includes(`${DEFAULT_MAX_LINES + 1} |`))
+			assert.ok(result.includes(`start_line=${DEFAULT_MAX_LINES + 1}`))
+		})
+
+		it("allows reading beyond default limit with explicit end_line", () => {
+			const endLine = DEFAULT_MAX_LINES + 200
+			const result = formatFileContentWithLineNumbers(bigContent, 1, endLine)
+			assert.ok(result.includes(`${endLine} | row${endLine}`))
+		})
+	})
+
+	describe("byte truncation interaction", () => {
+		it("preserves truncation notice without numbering it", () => {
+			const input =
+				"alpha\nbeta\n\n---\n\n[FILE TRUNCATED: This content is 1.0 MB but only the first 400 KB is shown (600 KB truncated).]"
+			const result = formatFileContentWithLineNumbers(input)
+			assert.ok(result.startsWith("1 | alpha\n2 | beta"))
+			assert.ok(
+				result.includes(
+					"[FILE TRUNCATED: This content is 1.0 MB but only the first 400 KB is shown (600 KB truncated).]",
+				),
+			)
+			assert.ok(!result.includes("3 |"))
+		})
+	})
+})
Author	SHA1	Message	Date
Robin Newhouse	c8e142d021	test: update unit test snapshots	2026-03-26 17:35:34 -07:00
Robin Newhouse	2616e9c9f6	fix: resolve duplicate step numbering in replace_in_file instructions Fixes the duplicate step 5 issue when NOTEBOOK_INSTRUCTIONS is concatenated with BASE_DIFF_INSTRUCTIONS for .ipynb files. Changes notebook instructions to step 6 to avoid ambiguity.	2026-03-26 12:18:13 -07:00
Robin Newhouse	d86762a991	feat: align read_file line format with SDK while keeping superior chunking - Change line format from 'L1:' to '1 \|' to match SDK format - Add proper NaN validation for start_line/end_line parameters - Keep 1000-line chunking with continuation hints (superior to SDK) - Update tool description and tests to reflect new format - Add directory usage guardrail back to tool description - Update replace_in_file prompt to reference new line format This aligns the PR with the newer SDK design patterns while preserving the superior chunking behavior that prevents context overflow issues.	2026-03-26 12:17:19 -07:00
Robin Newhouse	1ccd2505b8	feat(read_file): add chunked reading with start_line/end_line parameters Add optional start_line and end_line parameters to read_file so models can read files in chunks instead of loading entire files into context. Default limit is 1000 lines per read, with a continuation hint guiding the model to paginate when needed. Made-with: Cursor	2026-03-06 12:20:24 -08:00
Robin Newhouse	c70df9b1cf	Update prompt snapshots	2026-03-06 12:20:24 -08:00
Robin Newhouse	0c8b007b81	prompt: clarify read_file line labels for replace_in_file	2026-03-06 12:20:24 -08:00
Robin Newhouse	d1aa7bc0f8	fix(read_file): add stable line labels in act/plan	2026-03-06 12:20:24 -08:00