start-issuelisted

<!-- requires: coder, explorer --> <!-- chains: /test-issue --> <!-- generated from .ai/workflows/start-issue.md; do not edit directly --> > **Codex:** this skill is the `/start-issue` adapter. For chained workflows (/test-issue), invoke the matching generated skill (for example `heph:<workflow>`) when it is available; otherwise read and follow `.agents/skills/<workflow>/SKILL.md`. Use Codex role agents from `.codex/agents/` when the runtime exposes them; otherwise perform the work directly and keep the same structured output. > Codex does not substitute `$ARGUMENTS` — read it as the arguments given in the user's request. Start working on issue $ARGUMENTS. Run autonomously through the full plan-critique-implement cycle. ## Autonomy Rules - **Resolve ambiguity yourself.** If requirements are unclear, make reasonable assumptions based on codebase context, existing patterns, and the issue description. Document assumptions in commit messages and the eventual PR body. Do NOT stop to ask the user. - **Recover before escalating.** If an approach fails, try a different one. Only stop if you've exhausted alternatives AND the failure involves irreversible risk. ## Phase 1: Context 1. **Detect repo**: Run `git remote get-url origin` to identify the target repo. 2. **Load context in parallel**: - **Read the issue**: `gh issue view <#> --repo <detected-repo>` - **Explore codebase** (explorer role agent(s)): Search for relevant files, understand current implementation - **