claude-code - 💡(How to fix) Fix [Bug] claude-opus-4-6: Model behavior regression - hallucination and lost self-correction in code auditing [1 comments, 2 participants]

[{"error":"Error: NON-FATAL: Lock acquisition failed for /Users/fushihua/.local/share/claude/versions/2.1.104 (expected in multi-process scenarios)\n at Tc_ (/$bunfs/root/src/entrypoints/cli.js:2836:2153)\n at Se6 (/$bunfs/root/src/entrypoints/cli.js:2836:1233)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-13T00:33:14.722Z"},{"error":"Error: File does not exist. Note: your current working directory is /Users/fushihua/Desktop/CrabClaw.\n at call (/$bunfs/root/src/entrypoints/cli.js:4458:7632)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-13T00:34:26.885Z"},{"error":"MaxFileReadTokenExceededError: File content (23893 tokens) exceeds maximum allowed tokens (10000). Use offset and limit parameters to read specific portions of the file, or search for specific content instead of reading the whole file.\n at kY7 (/$bunfs/root/src/entrypoints/cli.js:4449:12671)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-13T00:34:53.982Z"},{"error":"MaxFileReadTokenExceededError: File content (12557 tokens) exceeds maximum allowed tokens (10000). Use offset and limit parameters to read specific portions of the file, or search for specific content instead of reading the whole file.\n at kY7 (/$bunfs/root/src/entrypoints/cli.js:4449:12671)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-13T00:36:19.859Z"},{"error":"Error: EISDIR: illegal operation on a directory, read '/Users/fushihua/Desktop/CrabClaw/CrabCode/crates/acosmi-supervisor/src'\n at G1H (/$bunfs/root/src/entrypoints/cli.js:3942:15399)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-13T00:36:29.205Z"},{"error":"MaxFileReadTokenExceededError: File content (31694 tokens) exceeds maximum allowed tokens (10000). Use offset and limit parameters to read specific portions of the file, or search for specific content instead of reading the whole file.\n at kY7 (/$bunfs/root/src/entrypoints/cli.js:4449:12671)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-13T00:36:34.275Z"},{"error":"MaxFileReadTokenExceededError: File content (11287 tokens) exceeds maximum allowed tokens (10000). Use offset and limit parameters to read specific portions of the file, or search for specific content instead of reading the whole file.\n …

TL;DR

The most likely fix involves addressing the MaxFileReadTokenExceededError by implementing a mechanism to read files in chunks or optimizing the file reading process to handle large files.

Guidance

1. Implement chunked file reading: Modify the code to read files in smaller chunks, using offset and limit parameters, to avoid exceeding the maximum allowed tokens.
2. Optimize file reading: Review the file reading process to optimize it for handling large files, potentially by searching for specific content instead of reading the whole file.
3. Error handling: Improve error handling to catch and handle MaxFileReadTokenExceededError and other errors, providing more informative error messages and potentially retrying the operation with adjusted parameters.
4. Model configuration: Consider adjusting the model's configuration, such as the maximum reasoning depth or the file reading limits, to better handle large files and complex tasks.
5. Verify model behavior: Thoroughly test the model's behavior after implementing these changes to ensure that it correctly handles file reading and auditing tasks.

Example

// Example of reading a file in chunks
{
  "readFile": {
    "filePath": "/path/to/file",
    "offset": 0,
    "limit": 10000
  }
}

Notes

The provided error logs indicate that the model is encountering issues with reading large files, which may be related to the MaxFileReadTokenExceededError. Addressing this error and optimizing the file reading process may help resolve the regression issues.

Recommendation

Apply a workaround by implementing chunked file reading and optimizing the file reading process to handle large files, as this is likely to address the root cause of the regression issues.