Why it matters
- Wrong files can crowd out the code that matters.
- Large prompts become harder to review and more expensive to run.
- Auditors need to know whether selected context fits the target model window before copy/export.
Token Budgeting
PastePrompt helps users understand context size before export by showing estimated token counts and model budget status while they select files.
How it works
- PastePrompt estimates tokens for files, folders, prompt instructions, file maps, and final bundles.
- A model budget selector lets users compare selected context against common context limits.
- Budget status shows whether the selection is under, near, or over the chosen limit.
Example workflow
- Select a repository and expand the file tree.
- Choose files or folders relevant to the review question.
- Set the target model budget.
- Remove low-value files or split the review if the selected context is too large.
- Generate the bundle only after the budget looks intentional.
Example UI description
- Selected files: 38
- Selected source tokens: 52,400
- Prompt instruction tokens: 740
- Estimated final bundle tokens: 58,900
- Budget selector: 64k
- Status: near budget
Limitations
- Token counts are estimates and can differ from provider-specific tokenization.
- Formatting overhead and model-specific message wrappers may change the final count.
- Users should leave margin when working near a model limit.
Build a repeatable context workflow.
Download the macOS app, review the docs, and contact support for Founder or Pro licensing during launch.