Dual Author — Data skill for Claude Code
Self-orchestrating GitHub issue pipeline — a Claude Code skill that implements, dual-reviews, and auto-merges issues inside herdr.
How to install Dual Author
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open Tudor0404/dual-author and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Dual Author does
Self-orchestrating GitHub issue pipeline — a Claude Code skill that implements, dual-reviews, and auto-merges issues inside herdr.
Alternatives in Data
- OpenViking — Self-evolving Context Database for AI Agents 33.3k ★
- Data Contract — This document defines which files belong to the system (auto-updatable) and which belong to the user (never to 32.5k ★
- Archify — Agent skill for beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams—self 17.2k ★
README
dual-author
A self-orchestrating **issue pipeline** that runs as a [Claude Code](https://claude.com/claude-code) skill inside [herdr](https://github.com/herdr).
Point it at one or more GitHub issues (or a project board, label, or milestone) and it will, for each issue, in parallel:
- Create a dedicated git worktree workspace.
- Implement the issue with a named Claude worker agent and push a draft PR.
- Dual-review the change with fresh
codex+claudereviewer agents in split panes, plus PR review bots (CodeRabbit, Copilot, CI). - Fix → re-review with fresh reviewer instances each round until the diff is clean (no round cap).
- Mark the PR ready and arm auto-merge — it merges only when all checks pass.
A live dashboard pane runs a full-screen [Textual](https://textual.textualize.io/) TUI: a selectable issues table (stage, timings, PR, checks), a per-issue detail panel (review-round verdicts, live worker output), an activity feed, and a `g` graph view: the blocking DAG between the run's issues (from GitHub issue dependencies) plus every issue's pipeline chain at once. It also live-renames each issue's workspace, agent, and tab (`# · `) so the whole herdr UI doubles as a status board. Falls back to a plain-text render without `uv`.
Requirements
- herdr — the skill must run inside a herdr-managed pane (
HERDR_ENV=1). - Claude Code (
claudeon yourPATH). - `codex` — used as the second reviewer.
- `gh` — authenticated (
gh auth login) withrepo,project, andworkflowscopes. python3— runs the monitoring/orchestration script.- `uv` (recommended) — provisions the TUI dashboard's env (python ≥3.10 + textual) on first use; without it the dashboard falls back to a plain-text render.
Install
Skills live in `~/.claude/skills//`. Clone this repo di
Related Skills
Google Maps Scraper MCP
Google Maps Scraper MCP — search businesses, extract reviews, emails, contacts and lead data from Google Maps
Sequential
Lanza el pipeline secuencial para multiples issues dentro de una sesion tmux. Cada issue se enruta automaticam
Self Eval
Self-improvement loop. After a pipeline run, reviews how faithfully it executed step-by-step (from the determi
Fix Tickets
Autonomously fix one or more speckit-companion GitHub issues with the SpecKit Companion pipeline — clean branc
Book Translation Skill
You are a book translation assistant. You translate entire books from one language to another by orchestrating
NLP Mir
Transformer-based NLP extraction pipeline achieving 95%+ accuracy on unstructured documents. Implements multi-
Related Agents
Profile Synthesizer
Synthesizes voice profiles from self-report scores and computational writing analysis. Merges dual outputs usi
Feature Implementer
Implement a ReviewFlow feature via TDD inside-out. Receives a validated plan (docs/plans/ .plan.md) and spec (
Worker Codex Implementation
Implements a ticket end-to-end inside an isolated worktree by orchestrating codex exec for code-gen. Commander