vector-store-stinger — Claude Code skill: vector storage design
vector-store-stinger is a skill for choosing pgvector, Deep Lake, or Qdrant, configuring indexes, and planning migrations for projects storing embeddings.
AGPL-3.0-or-later ★ 96 updated 2026-10-18
/plugin marketplace add legioncodeinc/vibe-coding-tools
/plugin install data-and-ai-tools@wasp-nestWhat vector-store-stinger does
vector-store-stinger covers the full lifecycle of vector data: greenfield store selection, schema design, index tuning, and migration planning. The skill references guides for pgvector with Drizzle (the default for this repository), Deep Lake as an option for existing projects, and Qdrant or managed services like Pinecone.
It enforces hard rules that prevent common mistakes: embedding dimension is a schema constraint, not a runtime detail; operator class must match the distance metric; every migration is broken into steps. For pgvector, separate guides cover Drizzle schemas, indexing, hybrid search (vector plus full-text), and migrations.
The skill does not handle security audits, credential storage, or embedding generation—those are delegated to other skills. PostgreSQL with the pgvector extension or your chosen alternative store is required to function.
What it helps with
- select a vector store for a new project with embeddings
- design a Drizzle schema with a vector column in PostgreSQL
- choose and configure HNSW or IVFFlat indexes
- plan and execute a migration when dimension or index type changes
- build hybrid search combining full-text and vector queries
Example prompts for the agent
- Help me pick a vector store: my project runs Postgres already, but I'm not sure whether pgvector or Qdrant fits better.
- Design a
documentstable with a 1536-dimensional vector column and recommend an index for fast retrieval. - Write a migration that adds hybrid search using both
tsvectorand vector columns with fusion ranking.
How to install vector-store-stinger
1Plugin marketplace
/plugin marketplace add legioncodeinc/vibe-coding-tools/plugin install data-and-ai-tools@wasp-nestThis installs the whole data-and-ai-tools plugin with all its skills
2Skills CLI (Vercel)
npx skills add legioncodeinc/vibe-coding-tools --skill vector-store-stinger -a claude-code3Manually: put the files in place
~/.claude/skills/vector-store-stinger/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/vector-store-stinger/
1Skills CLI (Vercel)
npx skills add legioncodeinc/vibe-coding-tools --skill vector-store-stinger -a codex2Manually: put the files in place
.agents/skills/vector-store-stinger/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add legioncodeinc/vibe-coding-tools --skill vector-store-stinger -a cursor2Manually: put the files in place
.cursor/skills/vector-store-stinger/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install legioncodeinc/vibe-coding-tools plugins/data-and-ai-tools/skills/vector-store-stinger2Manually: put the files in place
.github/skills/vector-store-stinger/SKILL.md1Skills CLI (Vercel)
npx skills add legioncodeinc/vibe-coding-tools --skill vector-store-stinger -a gemini-cli2Platform CLI
gemini skills install https://github.com/legioncodeinc/vibe-coding-tools --path plugins/data-and-ai-tools/skills/vector-store-stingerManually: put the files in place
.devin/skills/vector-store-stinger/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about vector-store-stinger
How do I install vector-store-stinger in Claude Code?
Run /plugin marketplace add legioncodeinc/vibe-coding-tools, then /plugin install data-and-ai-tools@wasp-nest, then npx skills add legioncodeinc/vibe-coding-tools --skill vector-store-stinger -a claude-code. Open ~/.claude/skills/vector-store-stinger/SKILL.md afterwards.
What does vector-store-stinger do?
It guides vector storage design: selecting the right store (pgvector, Deep Lake, Qdrant), building schemas with appropriate indexes, planning migrations, and structuring hybrid search that combines full-text and vector queries.
What is the default storage option?
Neon Postgres with the pgvector extension and Drizzle. Deep Lake, Qdrant, and managed services are documented as alternatives with different tradeoffs—the choice depends on your project's needs.