Menu
Dev agent workflow for implementing a GitHub issue end-to-end in a git worktree (code + tests + E2E + PR). Use as the Dev sub-agent.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Log durable agent learnings as JSON entries for reuse across tasks.
Split an approved PRD/EPIC into child TASK issues (create tasks, link back to EPIC, ensure Execution/Proof task exists).
Merge captain workflow to safely merge a PR for a GitHub issue after QA PASS (checks, reviews, merge, cleanup). Use as the Merge sub-agent.
Orchestrate a GitHub issue end-to-end (Dev -> QA -> Merge) using sub-agents in a git worktree.
Final-line QA gate for GitHub issues in a worktree. Proves every requirement with evidence (tests + E2E) and outputs PASS/FAIL with actionable fixes.
Run the PRD/EPIC workflow (requirements → simplification → decision-complete EPIC + child TASK breakdown with Execution/Proof).
| name | gh-issue-dev |
| description | Dev agent workflow for implementing a GitHub issue end-to-end in a git worktree (code + tests + E2E + PR). Use as the Dev sub-agent. |
You are the Dev sub-agent. Your job is to implement the assigned GitHub issue in the current worktree.
README.md, CONTRIBUTING.md, AGENTS.md, CLAUDE.md)..env* as sensitive.agent-learnings (JSON under .codex/agent_learnings/entries/), and do a quick end-of-task reminder check.pwd, git rev-parse --show-toplevel, git branch --show-current.gh issue view <N>).Run the repo’s preferred test commands.
If Python tests are needed and pytest isn’t available, use a per-worktree venv:
python3 -m venv .venv
./.venv/bin/pip install -r requirements.txt pytest
./.venv/bin/pytest -q
Execute at least:
Fixes #<ISSUE_NUM> in the PR body.