CtrlK
BlogDocsLog inGet started
Tessl Logo

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.

72

Quality

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Quality

Content

87%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, token-efficient reference skill with concrete tuning parameters and clean progressive disclosure to a real details file; its main gap is the absence of an explicit, checkpointed tuning workflow.

Suggestions

Add a short sequenced tuning workflow (e.g. 1. select index by data size, 2. start with default HNSW params, 3. benchmark with real queries, 4. validate recall, 5. re-tune) with an explicit validation checkpoint before applying changes.

Surface the recall-validation feedback loop as a checkpointed step rather than only a 'Do' bullet, since recall drift is the core failure mode for this skill.

Consider a one-line pointer to the specific worked example in references/details.md that matches the user's scenario (e.g. HNSW tuning vs. quantization) to improve navigation.

DimensionReasoningScore

Conciseness

The body is lean — compact tables and terse bullet lists with no explanation of basic vector-search concepts Claude already knows; every line earns its place.

3 / 3

Actionability

Concrete, actionable reference data is provided (M=16, efConstruction=100, efSearch=50 with effects; per-quantization byte budgets; index selection by data size). As an instruction/reference skill the absence of inline code is not penalized because the guidance is specific and actionable, with executable templates correctly deferred to references/details.md.

3 / 3

Workflow Clarity

Tuning is inherently a multi-step process (select index → set parameters → quantize → benchmark → validate recall), but the body presents reference tables and best-practice bullets rather than a sequenced workflow; benchmarking and recall monitoring appear only as loose Do's, not as explicit validation checkpoints, so it sits at the 'steps present but checkpoints missing/implicit' level rather than a fully sequenced flow.

2 / 3

Progressive Disclosure

The body is a clear overview that points to one-level-deep, well-signaled references ('Full template library and detailed worked examples live in references/details.md. Read that file when you need the concrete templates.'), and that referenced file exists, keeping content appropriately split.

3 / 3

Total

11

/

12

Passed

Description

90%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, concise description with an explicit Use-when trigger clause and natural domain keywords; its only weakness is a single abstract capability verb rather than an enumerated action list.

DimensionReasoningScore

Specificity

The capability statement 'Optimize vector index performance for latency, recall, and memory' names the domain and three concrete metrics, but relies on a single abstract verb ('Optimize') rather than listing multiple distinct actions, matching the anchor that names a domain and some actions but is not comprehensive.

2 / 3

Completeness

It explicitly answers both what ('Optimize vector index performance for latency, recall, and memory') and when via an explicit 'Use when tuning HNSW parameters...' trigger clause.

3 / 3

Trigger Term Quality

'tuning HNSW parameters, selecting quantization strategies, or scaling vector search infrastructure' covers the natural terms practitioners actually say (HNSW, quantization, vector search), giving good keyword coverage.

3 / 3

Distinctiveness Conflict Risk

The vector-index niche with HNSW/quantization/vector-search triggers is clearly distinct and unlikely to fire for unrelated skills.

3 / 3

Total

11

/

12

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
wshobson/agents
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.