Role:
Sole designer & engineer
Stack:
Go, JSON API
Status:
Live, free to use
Squish
Shrink a prompt before it ever reaches the model, and pay for fewer tokens. A deterministic four-stage pipeline compresses content for LLMs with no inference, no network calls and no second model in the loop: the same input always produces the same shorter output.
- Applied AI
- Go
- Deterministic
Four stages, no model in the loop
content → extract → synthesise → prioritise → adapt → compressed
Spike extractor
Scores every sentence by keyword salience and keeps the ones carrying the most meaning, dropping low-signal filler.
Context synthesiser
Removes duplication, shortens common phrases and abbreviates, tightening the wording without changing what it says.
Token prioritiser
Enforces a maxTokens budget, trimming the lowest-priority content first so the result fits inside the limit you set.
Multi-AI adapter
Adds model-specific framing for the target (Claude, GPT or Cursor), so the compressed prompt arrives in a shape each model reads well.
Web app & API
Open the web UI, paste some content, pick a compression level (conservative, balanced, aggressive), a target model and a token budget, and watch the compressed output appear with its metrics and a ready-to-run curl command. Or hit the API directly: POST /api/optimise for a single prompt, POST /api/batch for up to fifty at once, GET /api/stats for supported models and levels.