What problem does it solve? GitHub Issue operations performed by AI agents can silently corrupt content through encoding errors, broken Markdown tables, leftover placeholders, or one-directional parent/child links. This Skill supplies verified operational procedures that prevent these failures during Issue creation, update, and linking. ## Core Features & Use Cases - Safe Issue Operations: Standardized call format for issue_create, issue_update, and comment_create via the agentdev_gh custom tool, including double-quote avoidance and explicit labels arguments. - Post-Write Verification: Read-back comparison of written content covering encoding, Markdown structure, required template sections, and repository reference links, with a three-stage retry policy. - Parent/Child Link Validation: Bidirectional verification of Parent: #N links between Epic and child Issues, plus Epic status tracking table updates with single-writer constraints. - Use Case: When a case-open command creates an Epic Issue and child Issues, use this Skill to confirm each Issue body was written correctly, replace #{TBD} placeholders with real numbers, and verify both directions of the parent/child links before proceeding. ## Quick Start Ask the agent to create a GitHub Issue and verify the written content, parent/child links, and Epic status table using the safe Issue operation procedures.
