Enforces line budgets for context-heavy files to ensure fast loading and optimal LLM performance.
Size Guard is an essential utility for maintaining efficient AI workflows by preventing context-heavy files from becoming bloated and slow. It monitors your project's 'memory' and documentation files against predefined limits set in a YAML configuration, automatically rotating or splitting files that exceed their line budgets. By keeping context chunks manageable and updating indexes to reflect new file structures, it ensures Claude can process information quickly without hitting context window limitations or experiencing latency, all while preserving historical data through secure archiving.
Key Features
01Intelligent file rotation and splitting for context optimization
02Automatic index updates to maintain file traceability
03Automated line budget enforcement via YAML configuration
04History preservation through secure archiving guardrails
05Transparent action logging for all file modifications
060 GitHub stars
Use Cases
01Managing large agent-memory files in long-running quantitative research projects
02Preventing LLM context window overflow by automatically chunking documentation
03Maintaining workspace performance by rotating verbose log and history files
