Paste raw dev
output.
Spend fewer
tokens in Codex.
Drop in terminal, VS Code, or CI output. tokensift removes token-eating junk before you send a root-cause brief to Codex, Claude Code, or another coding agent.
Raw paste → token-light prompt.
Review before sending.
- Detected type
- ex) CUDA / PyTorch OOM
- Input characters
- ex) 683
- Output characters
- ex) 561
- Input tokens
- ex) 172 approx.
- Output tokens
- ex) 133 approx.
- Reduction
- ex) 18%
- Noise removed summary
- ex) progress, wandb, NCCL, and worker INFO folded
- ex) torch dependency internals folded
FAQ
Common questions
How do I use it?
Paste raw terminal, VS Code, or CI output, click Sift & copy, review the compact Markdown brief, then send it to your LLM or coding agent.
What happens to my logs?
Raw logs are not sent to analytics or ads. Review the cleaned prompt before sharing it with any LLM or teammate.
What should I paste?
Paste the raw failure output you would normally copy from iTerm2, VS Code, CI, Docker, pytest, npm, or a training run.
What does it preserve?
The primary error, command, environment lines, user frames, memory/type/assertion/build details, and a final debugging question.
What does it fold?
Token-eating junk: progress bars, repeated warnings, timestamps, dependency frames, cached layers, wandb/NCCL spam, and noisy build or CI output.
Does it fix code automatically?
No. It prepares a cleaner prompt for ChatGPT, Claude Code, Codex, OpenCode, or another LLM.
Coverage
Raw outputs tokensift understands
Privacy-aware, prompt-first
tokensift focuses on trimming token-eating junk from raw development errors before you hand them to a coding agent. Analytics and ads are disabled by default; if enabled later, only aggregate events are allowed. Feedback: contact@tokensift.app.