claude-code - 💡(How to fix) Fix [Bug] Explore agent fails with zero tool uses at conversation start [2 comments, 3 participants]

[{"error":"Error: NON-FATAL: Lock acquisition failed for /Users/tory/.local/share/claude/versions/2.1.123 (expected in multi-process scenarios)\n at x86 (/$bunfs/root/src/entrypoints/cli.js:2769:2177)\n at Jw8 (/$bunfs/root/src/entrypoints/cli.js:2769:1257)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T13:36:26.855Z"},{"error":"Error: 400 {"type":"error","error":{"type":"invalid_request_error","message":"prompt is too long: 213605 tokens > 200000 maximum"},"request_id":"req_011Caa47XwG76z1iWFPoDq6K"}\n at generate (/$bunfs/root/src/entrypoints/cli.js:11:57693)\n at makeRequest (/$bunfs/root/src/entrypoints/cli.js:50:4943)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T13:43:33.830Z"},{"error":"Error: 400 {"type":"error","error":{"type":"invalid_request_error","message":"prompt is too long: 213605 tokens > 200000 maximum"},"request_id":"req_011Caa48CQnjJ6MrRLeachxB"}\n at generate (/$bunfs/root/src/entrypoints/cli.js:11:57693)\n at makeRequest (/$bunfs/root/src/entrypoints/cli.js:50:4943)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T13:43:43.248Z"},{"error":"Error: 400 {"type":"error","error":{"type":"invalid_request_error","message":"prompt is too long: 213605 tokens > 200000 maximum"},"request_id":"req_011Caa48pMkkrYpJAThpkano"}\n at generate (/$bunfs/root/src/entrypoints/cli.js:11:57693)\n at makeRequest (/$bunfs/root/src/entrypoints/cli.js:50:4943)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-04-30T13:43:51.297Z"}]

TL;DR

The issue is likely caused by the prompt exceeding the maximum token limit of 200000, resulting in an "invalid_request_error".

Guidance

• The error messages indicate that the prompt is too long, with 213605 tokens exceeding the 200000 maximum, suggesting that the input needs to be truncated or optimized.
• Review the conversation start context to identify potential areas where the prompt can be shortened without losing essential information.
• Consider implementing a prompt truncation mechanism to ensure it stays within the allowed limit.
• Verify that the issue persists after applying any changes to the prompt or input processing.

Example

No code snippet is provided as the issue seems related to input data rather than code.

Notes

The provided errors and environment information do not point to a specific code issue but rather a limitation or constraint on the input size. The solution will likely involve adjusting the input or how it's processed rather than changing the code itself.

Recommendation

Apply workaround: Truncate or optimize the prompt to stay within the 200000 token limit to prevent the "invalid_request_error".