From a focused fix to a project audit: explore engineering examples and glyph usage. This page includes a measurement from the team’s work and estimates based on Sol, Terra and Luna runs.
These examples use benchmark tasks and recorded glyph usage from Sol, Terra and Luna. They provide a reference for similar, bounded work with a verifiable outcome, rather than a fixed-price promise.
A ledger service needs a new operation: validate amounts, apply transfers atomically and handle retries and simultaneous requests correctly.
Expected resultA contract-compliant implementation, preserved balances and regression tests for failures and replay.
The service sometimes uses stale permissions after access is revoked. Check the cache, concurrent requests, cancellation and user isolation.
Expected resultRegression tests that detect the faults and a fix for the cache behavior.
A new snapshot format must read older data, preserve operation history and handle repeated migration safely.
Expected resultA compatible migration with metadata preservation and tests for conflicts, retries and failures.
Event processing has stopped. Use incident evidence and the operations runbook to determine a safe recovery sequence.
Expected resultAn executable recovery plan validated by simulation, without skipped events or duplicate effects.
Usage comes from glyphs recorded by the engine during execution, rounded up to whole glyphs. Unsuccessful work is also shown and marked. These estimates describe controlled examples: your project size, available context, mode and follow-up fixes affect the total. Check glyph history for actual account charges.
Our team observed an audit of a project with roughly 2500 files using about 130 glyphs. Usage depends on context, mode and actions. Project size does not imply every file was reviewed; fixes and repeat checks are separate tasks.
A security audit of a project containing about 2500 files used approximately 130 glyphs, as observed by the team. This describes a particular run, rather than a fixed price for auditing any project.
Project size describes the context and does not mean every file was inspected. Findings require review and prioritization. Fixes and follow-up verification are subsequent tasks with separate usage.
Start with a defined scope: a module, bug or screen. Compare the delivered result with the usage recorded in glyph history. Context size, mode, tools and duration all affect consumption.
Download Desktop, connect your project and start with a task whose result you can verify.