SQLite est simple, mais son utilisation en production exige des modèles sûrs pour les connexions, les schémas, les migrations et l'asynchrone. Cette compétence fournit à Claude, Codex et Claude Code des modèles pratiques pour la persistance locale.

Les données de configuration structurées peuvent être difficiles à inspecter dans des fichiers JSON, YAML et TOML. Cette compétence guide les requêtes jq et yq pour l’extraction, le filtrage, la conversion et les mises à jour.

Profile research can require several lookups and careful spending control. This skill guides approved Datacircle enrichment, summarizes profiles, and reports costs and balances.

Financial questions require selecting relevant records and interpreting amounts, permissions, and data freshness. This skill guides agents through authorized Link queries and clear financial summaries.

Customer reviews contain useful feedback that can be difficult to organize. This skill groups retrieved Google reviews into evidenced themes and suggests practical business improvements.

Comparing branch performance is difficult when reports use different periods and formats. This skill organizes Reach data into one consistent table with changes and attention items.

Weekly listing activity, reviews, and rankings can be difficult to compare consistently. This skill uses Reach MCP to produce a fixed, read-only report with changes and data notes.

Streaming analytics requires coordinated ingestion, table management, and query configuration. This skill provides Cloudflare Basin references and implementation patterns across Pipelines, Catalog, and SQL.

Les historiques de sessions de codage IA sont difficiles à examiner sous forme JSONL brute. Cette compétence convertit les sessions Claude Code et Codex en markdown lisible et en index consultables.

Survey totals can hide differences between recruitment channels. This skill guides your assistant to compare tagged WebAsk responses by volume, completion rate, and your selected outcome.

Sharing survey results requires choosing the right format and checking what recipients can see. This skill guides WebAsk tools to prepare links, exports, and AI reports with privacy checks.

Survey totals and raw responses can make conclusions difficult to identify. This skill guides assistants through WebAsk summaries, filtered reports, and text themes to explain results clearly.

Low survey completion can leave teams unsure which questions need attention. This skill uses WebAsk reports and questionnaire structure to identify potential drop-off points and recommend targeted changes.

Survey response drops are difficult to explain without knowing when the survey changed. This skill compares response trends with publication history and checks integrations.

Open survey responses are difficult to compare as long lists of comments. This skill organizes WebAsk responses into themes with counts, percentages, and representative quotes.

Survey completion counts can hide missing or unusable contact details. This skill guides assistants through WebAsk reports to assess contact quality, compare channels, and prioritize improvements.

Suspicious survey responses can distort results and weaken decisions. This skill guides checks of WebAsk responses and compares key metrics before recommending exclusions.

Survey scores can mislead when questionnaire changes, sample sizes, and seasonal differences are ignored. This skill guides WebAsk comparisons using version history, response distributions, and shared survey metrics.

Test submissions and duplicate responses can distort survey reports and complicate follow-up. This skill guides preview-based WebAsk cleanup with reversible hiding, response tags, and explicit approval before deletion.

Small survey samples can make random differences look like meaningful improvements. This skill guides comparisons using outcome counts, percentage-point differences, and checks for comparable groups.