generated: '2026-08-29' method: searched source: >- Two skills are PROVIDER-PUBLISHED and saved verbatim from https://github.com/qlik-oss/agentic-skills — Qlik's own public Agent Skills hub, which validates every contribution in CI against the agentskills.io SKILL.md specification and ships them as a Claude Code plugin marketplace. Three more are generated by API Evangelist and grounded in operationIds verified verbatim in openapi/*.json. description: >- Packaged Agent Skills for Qlik Cloud. Qlik is one of the few providers in the catalog that publishes Agent Skills itself rather than leaving them to be derived — official/ skills are Qlik-maintained, community/ skills pass a security scan and review before merge. provider_hub: repo: https://github.com/qlik-oss/agentic-skills license: Apache-2.0 specification: https://agentskills.io/specification install: npx skills add qlik-oss/agentic-skills claude_plugins: - qlik-cloud-skills@qlik-cloud-skills - qlik-cloud-community-skills@qlik-cloud-skills tiers: [official, community] note: >- Each provider skill ships a references/ directory alongside its SKILL.md in the source repo. Only the SKILL.md is mirrored here; the references were not copied. skills: - file: qliksense-ai-readiness-optimizer.md name: Qlik AI Readiness Optimizer tier: official method: searched upstream: https://github.com/qlik-oss/agentic-skills/blob/main/official/skills/qlik-ai-readiness-optimizer/SKILL.md author: Qlik (JoshQlikDesign) version: 2.0.0 license: Apache-2.0 summary: >- Analyzes a Qlik app over MCP and optimizes it for AI consumption (Qlik Answers and the Qlik MCP server) against a 5-layer model — field naming, grouping, master-item descriptions, field noise, date fields, synonyms. - file: qliksense-load-script.md name: Qlik Load Script Generator tier: community method: searched upstream: https://github.com/qlik-oss/agentic-skills/blob/main/community/skills/qlik-load-script/SKILL.md author: nabeel-oz version: 1.0.0 license: Apache-2.0 summary: >- Writes and extends Qlik Sense load scripts (.qvs), with a focus on preparing flat datasets for Qlik Predict ML experiments. - file: qliksense-onboard-user-to-space.md name: Onboard a user into a Qlik Cloud space tier: api-evangelist method: generated api: openapi/qliksense-users.json, openapi/qliksense-spaces.json operations: [inviteUsers, getUsers, getSpaces, createSpace, getTypes, createSpaceAssignment, getSpaceAssignments] - file: qliksense-reload-an-app.md name: Reload a Qlik analytics app tier: api-evangelist method: generated api: openapi/qliksense-scheduling-tasks.json, openapi/qliksense-reloads.json operations: [listTasks, createTask, startTask, getTaskRuns, getTaskRunLog, getTaskLastRun] - file: qliksense-subscribe-to-events.md name: Subscribe to Qlik Cloud events tier: api-evangelist method: generated api: openapi/qliksense-webhooks.json operations: [listEventTypes, listWebhookEntries, createWebhook, getWebhookDeliveryList, getWebhookDelivery, resendDelivery, deleteWebhook] grounding_caveat: >- Only 504 of the 681 published operations carry an operationId. Where a needed operation has none — the whole Reloads API, most of the Apps API — the generated skills cite METHOD + PATH instead. No operationId is invented.