Switching between AI work and Linear slows issue tracking. This skill guides Claude, Codex, and Claude Code through Linear GraphQL operations.

Automatic model selection can pick a different model than the user intended. This skill lets users explicitly route a prompt to Haiku, Sonnet, or Opus.

Accumulated learning files can become stale or noisy over time. This skill guides Claude, Codex, and Claude Code through a confirmed reset that preserves the file structure.

Long AI sessions can lose useful decisions and patterns. This skill turns on automatic insight capture into a local knowledge base.

Continuous learning can keep extracting insights after a focused session ends. This skill turns it off and confirms the learning state clearly.

Project knowledge bases can become hard to inspect during active work. This skill shows learning status, entry counts, recent insights, and cache statistics in one summary.

Complex coding requests can drift when work begins too soon. This skill helps Claude and Codex create staged plans with questions and verification.

Choosing between many skills and agents can delay technical work. This skill maps each task to focused tools using concise catalogs and selection rules.

AI coding sessions can lose track of project plans when notes are stale or missing. This skill checks plan status and suggests the right save or sync action.

SAFe planning in Jira is hard to keep consistent across epics, stories, and subtasks. This skill provides structured templates and scripts for Jira issue creation and workflow updates.

Jira updates often drift from development work and require manual follow-up. This skill helps Claude, Codex, or Claude Code propose stories, request approval, create issues, and sync workflow status.