Tune memory harness recall, decay-sweep, and write batching parameters.
npx skills add https://github.com/AesopScott/mojo --skill index-tuner
Please help me install this Agent Skill. Skill: index-tuner Source: https://github.com/AesopScott/mojo/tree/main/harnesses/skills/index-tuner Command: npx skills add https://github.com/AesopScott/mojo --skill index-tuner
This Skill addresses performance bottlenecks in the Memory harness by providing a structured approach to tuning recall, decay, and batching parameters to ensure system responsiveness as data grows.
Use the index-tuner skill to inspect the current Memory harness configuration and propose a performance optimization for the recall index.
High-intent search queries and answers about installing and using this skill.
To reduce recall latency, tune memory harness parameters like the recall index, decay-sweep intervals, and write batching to systematically target bottlenecks and restore system throughput as data grows.
Memory write batching improves performance by grouping write operations to reduce processing overhead and latency, functioning as a tunable parameter within the memory harness to maintain high throughput for growing datasets.
Adjust decay-sweep intervals by inspecting the current memory harness configuration and modifying the parameter boundaries, verifying changes through positive and boundary testing to ensure optimized responsiveness.
Yes, performance tuning applies to existing memory management systems to fix increasing latency, providing a standardized runbook to inspect and modify the memory harness boundary safely without disruption.
The best way to tune recall index parameters safely is to follow defined safety protocols and verify configuration changes through positive and boundary testing before applying them to the live memory harness.
Memory recall latency increases with a growing footprint due to performance bottlenecks in the memory harness, requiring targeted tuning of recall, decay, and batching parameters to restore system responsiveness.
EngineerWithAITune HNSW parameters and quantization strategies for Qdrant vector indexes.
AesopScottImplement transport adapters and node-trust policies for network protocol configurations.
JhabbigTune HNSW parameters, quantization, and Qdrant collection settings for vector search performance.
emilneuraz-aiTune HNSW parameters and quantization to optimize vector index latency and recall.
