LLM evaluation is hard to make consistent across teams and models. This skill helps design judge prompts, rubrics, metrics, and bias checks for repeatable assessments.

Writing OQL queries requires exact schema names, JSON types, Candid escaping, and careful pagination. This skill provides practical patterns for reliable, read-only canister queries.

Logseq database concepts can be difficult to connect across entities, classes, properties, and validation. This skill explains the Datascript model and provides practical schema templates.

Logseq users often need reliable access to pages, blocks, properties, and relationships without hand-writing every query. This skill gives Claude, Codex, and Claude Code structured helpers for reading local Logseq graph data.

Logseq database queries are difficult to structure and optimize. This skill provides practical Datalog patterns for filters, pulls, rules, aggregations, and plugin queries.

Logseq graph migrations can lose structure, types, links, and query behavior. This skill provides a staged plan for backup, import decisions, and validation.

Complex Logseq templates can hide structural errors and uneven modules. This skill analyzes relationships, distributions, quality, and differences in clear reports.

Manual collection of YouTube channel metrics is slow and inconsistent. This skill runs a BrowserAct workflow and returns structured channel and video details.

Collecting transcripts from multiple YouTube videos takes time and custom scraping. This skill uses BrowserAct to return transcripts and related metadata from keyword searches.

Collecting comparable local business data manually is slow and inconsistent. This skill submits Google Maps searches through BrowserAct and monitors each task.

Manual local business research is slow and difficult to standardize. This skill submits configurable Google Maps searches to BrowserAct and monitors workflow progress.

Manual review collection is slow and difficult to monitor across locations. This skill starts BrowserAct review workflows and reports task progress from the command line.

Collecting Amazon review details manually is slow. This skill submits an ASIN to BrowserAct and monitors a hosted review extraction workflow.

Amazon sellers need current Buy Box, price, fulfillment, and competitor information for specific listings. This skill submits ASINs to BrowserAct and returns structured marketplace results.

Manual Amazon product checks are slow and inconsistent. This skill uses BrowserAct to retrieve structured product details for a supplied ASIN.

Data pipeline work can become inconsistent when ingestion, features, quality, and reporting are planned separately. This skill organizes data tasks around pipeline context, quality gates, lineage, and handoffs.

Desmos blocks can fail when settings, equations, and restrictions use the wrong syntax. This skill creates parser-aware graph blocks with reliable styling and bounds.

Database work often requires exact psql commands and connection settings. This skill provides invocation patterns for queries, meta-commands, SQL files, and output options.

BigQuery exploration can expose sensitive data or consume excessive resources. This skill provides cautious bq patterns for schema inspection, table discovery, row counts, and cost checks.

Data teams need a safer way to load S3 files into Bauplan without changing main immediately. This skill guides a branch-based Write-Audit-Publish flow with checks before publish.