Orchestrates issue creation with automatic review. Use when asked to "create and review issues", "create verified issues", or when you want issues automatically validated against requirements.
View on GitHublangadventurellc/claude-marketplace
task-trellis
plugins/task-trellis/skills/issue-creation-orchestration/SKILL.md
February 4, 2026
Select agents to install to:
npx add-skill https://github.com/langadventurellc/claude-marketplace/blob/main/plugins/task-trellis/skills/issue-creation-orchestration/SKILL.md -a claude-code --skill issue-creation-orchestrationInstallation paths:
.claude/skills/issue-creation-orchestration/# Issue Creation with Review Orchestrate issue creation with automatic review to ensure created issues accurately reflect the original requirements. ## Goal Create Trellis issues using the `issue-creation` skill, then automatically verify them against the original requirements using `issue-creation-review`. Handle any questions or findings from the review before completing. ## Critical Rule: Create Only Immediate Children (One Level Down) **When given a parent issue, create only its immediate children - ONE level down in the hierarchy.** - For a **Project**: Create only Epics, then STOP - For an **Epic**: Create only Features, then STOP - For a **Feature**: Create only Tasks, then STOP **Do NOT recursively decompose further.** If the user gives you a Project, create the Epics and stop. Do not continue to create Features or Tasks. The user will invoke this skill again on individual children if they want further decomposition. ## Autonomous Operation **When given a parent issue ID** (e.g., "F-feature-id", "E-epic-id"), proceed directly to creating the immediate child issues (one level down) without asking for confirmation. The user has already decided they want child issues created by invoking this skill. **Do not ask about granularity.** Default to coarser-grained issues - meaning fewer, larger issues at the level you're creating. This does NOT mean decomposing further down the hierarchy. Use your judgment to determine appropriate scope at the current level. **Only ask clarifying questions when:** - Requirements are genuinely ambiguous and could be interpreted multiple ways - Critical information is missing that cannot be inferred from context - A decision has significant irreversible consequences Otherwise, make reasonable decisions and proceed. ## Input `$ARGUMENTS` - The user's original requirements/instructions for issue creation ## Process ### 1. Capture Original Input **CRITICAL**: Store the exact user instructions verbatim at the start. ``` O