spfresh-reviewer

Audit SPFresh vector index code against SPANN and SPFresh paper specifications.

3|Updated Aug 4, 2025
One-click install
npx skills add https://github.com/birdayz/fdb-go --skill spfresh-reviewer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spfresh-reviewer
Source: https://github.com/birdayz/fdb-go/tree/main/.claude/skills/spfresh-reviewer
Command: npx skills add https://github.com/birdayz/fdb-go --skill spfresh-reviewer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill ensures the SPFresh vector index implementation is in accordance with its algorithmic specifications provided in the SPANN and SPFresh papers.

Core Features & Use Cases

  • Algorithmic Compliance Check: Reviews SPFresh code against the SPANN and SPFresh papers.
  • Recall/Latency Audits: Audits for recall and latency regressions.
  • Periodic Audits: Schedules regular checks for faithfulness to the paper specification.

Quick Start

Review the SPFresh code changes with the SPFresh Paper Review skill, specifying the updated implementation code or a benchmark run.

Frequently Asked Questions about spfresh-reviewer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit an FDB-native SPFresh implementation for algorithmic fidelity?

To audit an FDB-native SPFresh implementation, review the code against the SPANN and SPFresh papers to validate posting list invariants, closure replication, query-aware pruning, the LIRE protocol, and two-level routing for algorithmic fidelity.

What is the LIRE protocol in SPFresh and how do I check if my code adheres to it?

The LIRE protocol is a core algorithmic specification in SPFresh. You can check code adherence by reviewing the implementation logic against the SPFresh paper to ensure the protocol's replication and routing invariants are correctly maintained.

Can I use a code audit to diagnose recall and latency regressions in a SPFresh vector index?

Yes, you can diagnose recall and latency regressions by auditing the SPFresh vector index code against its algorithmic specifications, ensuring that posting list invariants and query-aware pruning mechanisms function correctly.

How do I verify that SPFresh posting list invariants and two-level routing are correctly implemented?

Verify posting list invariants and two-level routing by auditing the SPFresh code changes against the SPANN paper's specifications, checking that the dynamic posting list maintenance and routing logic match the algorithmic design.

Does this SPANN algorithmic compliance check support periodic audits of vector index changes?

Yes, the SPFresh algorithmic compliance check supports periodic audits by scheduling regular reviews of the vector index implementation to ensure continued faithfulness to the paper specification over time.