vector-index-tuning
Optimize vector index performance for latency, recall, and memory. Use when tuning HNSW parameters, selecting quantization strategies, or scaling vector search infrastructure.
80
71%
Does it follow best practices?
Impact
100%
1.56xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./tests/ext_conformance/artifacts/agents-wshobson/llm-application-dev/skills/vector-index-tuning/SKILL.mdQuality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that concisely covers specific capabilities, includes domain-appropriate trigger terms, and clearly delineates both what the skill does and when to use it. The description uses proper third-person voice and targets a well-defined niche that minimizes conflict risk with other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'optimize vector index performance', 'tuning HNSW parameters', 'selecting quantization strategies', 'scaling vector search infrastructure'. These are concrete, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers both 'what' (optimize vector index performance for latency, recall, and memory) and 'when' (explicit 'Use when' clause covering tuning HNSW parameters, selecting quantization strategies, or scaling vector search infrastructure). | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords that a user working in this domain would use: 'vector index', 'latency', 'recall', 'memory', 'HNSW parameters', 'quantization', 'vector search'. These are the exact terms practitioners would mention. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche — vector index optimization with specific triggers like HNSW, quantization strategies, and vector search infrastructure are unlikely to conflict with other skills. This is a very specialized domain. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides highly actionable, executable code templates for vector index tuning but is severely bloated—most of the code is boilerplate that Claude can generate from brief instructions. The high-value content (index selection table, HNSW parameter table, quantization comparison) is buried among hundreds of lines of template code. It lacks a clear tuning workflow with validation steps and feedback loops.
Suggestions
Reduce to ~50-80 lines: keep the decision tables and parameter recommendations inline, move all code templates to separate files (e.g., templates/hnsw_benchmark.py) with one-line descriptions and links.
Add an explicit tuning workflow with validation: e.g., '1. Benchmark baseline → 2. Adjust M/ef → 3. Measure recall@k → 4. If recall < target, increase ef_search → 5. Apply quantization → 6. Re-benchmark to verify recall held'.
Remove code that Claude can trivially generate (scalar quantization, binary quantization, basic benchmarking loops) and replace with brief descriptions of when to use each approach.
Add a decision flowchart or checklist for choosing between optimization strategies (recall vs speed vs memory) rather than presenting all configurations without guidance on selection.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~400+ lines, with massive code templates that Claude could generate on its own. The quantization class, monitoring code, and Qdrant configuration are all things Claude already knows how to write. The reference tables (index selection, HNSW parameters, quantization types) are the only high-value-density content. | 1 / 3 |
Actionability | The code templates are fully executable with concrete implementations using real libraries (hnswlib, qdrant_client, sklearn). Functions include proper type hints, parameters, and return values. The benchmarking and monitoring code is copy-paste ready. | 3 / 3 |
Workflow Clarity | While individual templates are clear, there's no overarching workflow connecting them (e.g., 'first benchmark, then select parameters, then quantize, then monitor'). There are no validation checkpoints or feedback loops for the tuning process—no guidance on what to do when recall drops or latency exceeds targets. | 2 / 3 |
Progressive Disclosure | This is a monolithic wall of code. The four large templates should be in separate reference files, with SKILL.md containing only the decision tables, parameter recommendations, and links to template files. The Resources section links externally but doesn't organize internal content across files. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (524 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- Dicklesworthstone/pi_agent_rust
- Commit
6e3d68c
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.