localnest: Skill for Claude Code

.claude/skills/localnest-sql-adapter/SKILL.md

localnest-sql-adapter is a skill for Claude Code from wmt-mobile/localnest. It costs 39 tokens per session (423 once invoked), scanned A, original, MIT.

A project-specific guide for changing LocalNest code that uses SQLite, a file-based database, for indexing and memory. It covers database adapter boundaries, migrations, backend selection, and sqlite-vec, an extension for vector search.

In plain words
What is it for?
Use it when changing database connections, schemas, migrations, extension loading, backend fallbacks, or retrieval and memory tests in LocalNest.
Why use it?
It helps prevent database refactors from changing query behavior or breaking retrieval and memory features.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter.

This is wmt-mobile/localnest's own configuration. It tells Claude Code how to work on localnest itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything localnest configures →

Reuse

Borrowing it

Nothing to install: this file belongs to wmt-mobile/localnest. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/wmt-mobile/localnest/main/.claude/skills/localnest-sql-adapter/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/wmt-mobile/localnest

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for localnest-sql-adapter

README.md
[![agentmods](https://agentmods.dev/badge/skills/wmt-mobile/localnest/localnest-sql-adapter.svg)](https://agentmods.dev/skills/wmt-mobile/localnest/localnest-sql-adapter)
Your own site
<a href="https://agentmods.dev/skills/wmt-mobile/localnest/localnest-sql-adapter"><img src="https://agentmods.dev/badge/skills/wmt-mobile/localnest/localnest-sql-adapter.svg" alt="Measured on agentmods" height="20"></a>
Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 423 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00039 $0.00423
Opus 5 $0.00019 $0.00211
Sonnet 5 $0.00008 $0.00085
Haiku 4.5 $0.00004 $0.00042

Measured 7d ago against content hash 932c724c732a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

localnest-sql-adapter scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/localnest-sql-adapter/SKILL.md · 66 lines

What it actually says

LocalNest SQL Adapter

Use this skill when working on LocalNest database access in:

  • src/services/retrieval/sqlite-vec/
  • src/services/memory/
  • src/app/create-services.ts
  • runtime/setup code that selects SQLite backends

Quick Start

  1. Find direct DB bindings first.
  2. Separate adapter-selection changes from query/schema changes.
  3. Preserve current Node 22 behavior unless the task explicitly changes defaults.
  4. Test retrieval and memory paths independently.

Workflow

1. Map the current DB entry points

Check for:

  • direct imports of node:sqlite
  • direct DB constructors
  • extension loading calls
  • transaction wrappers
  • schema and migration helpers

2. Contain the blast radius

When introducing a new adapter or backend path:

  • keep query semantics unchanged first
  • avoid mixing schema edits with adapter refactors
  • keep the active default path identical for current users until compatibility is proven

3. Verify extension-sensitive behavior

For sqlite-vec work, confirm:

  • connection creation allows extension loading
  • vec0 path resolution still works
  • status output reports whether the extension actually loaded

4. Test both data planes

Do not stop at retrieval tests if memory also touches SQLite. At minimum, cover:

  • index creation
  • re-index and update behavior
  • semantic search fallback behavior
  • memory backend detection and status

Guardrails

  • Do not eagerly import node:sqlite from general-purpose barrel modules.
  • Prefer lazy imports for sqlite-specific code paths.
  • If adding compatibility adapters, keep them behind explicit selection until verified.
  • Surface adapter and backend choice in status output so failures are diagnosable.

Reference

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 7d ago First seen · 66 lines · 39 tokens per session scan A 932c724c732a

Subscribe to this mod's changes

localnest-sql-adapter is a skill published in the GitHub repository wmt-mobile/localnest (6 stars, last pushed 2mo ago), licensed MIT. It adds 39 tokens to every session and 423 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

orchardcore-ai-memory-elasticsearch

Skill for indexing CrestApps Orchard Core AI Memory with Elasticsearch. Covers memory field mappings, dense vectors, k-nearest-neighbor retrieval, default query fields, user isolation, and Elasticsearch operations. Use this skill when requests mention AI Memory Elasticsearch, AIMemoryElasticsearchIndexProfileHandler…

CrestApps/CrestApps.AgentSkills · 130 tokens

crestapps-core-ai-memory

Skill for durable user memory, safety validation, and memory-aware orchestration in CrestApps.Core.

CrestApps/CrestApps.AgentSkills · 25 tokens

narsil

Use narsil-mcp code intelligence tools effectively. Use when searching code, finding symbols, analyzing call graphs, scanning for security vulnerabilities, exploring dependencies, or performing static analysis on indexed repositories.

postrv/narsil-mcp · 42 tokens

setup-location-memory

Install or repair the location archive. Use when the location tools are missing, when they error on the database, when the user wants to import a Google Timeline export, when they want to start a live feed from OwnTracks, or when they ask to set up location memory.

michelgrolet/tars-location-mcp · 59 tokens

lore-onboard

First-time setup wizard for LoreConvo. Verifies the MCP server is connected, the database is accessible, hooks are configured, and runs a test save/load cycle. Use when the user says "set up loreconvo", "onboard", "/lore-onboard", "verify loreconvo", "test loreconvo setup", or after first installing the plugin.

labyrinth-analytics/loreconvo · 81 tokens

sepia

Use when the user's AI assistant should recall or persist long-term knowledge about the user, their projects, preferences, decisions, people, conventions, or technical stack — across sessions and across tools. Triggers: starting meaningful work ("remember that", "what do we know about", "recall", "save this for…

Michael-Obele/sepia · 99 tokens