claudecodex

tmux-pair

Run a single coding agent on a task via tmux + git worktree. /run auto-entry performs a short repo + task recon, then dispatches to /solo. /solo runs one agent in a fresh worktree with a gated 7-phase workflow (Recon, Clarify, Reviewer-Readiness, Plan-Check, Implementation + bullet commits, Final-Verify, Persist, Commit hygiene, Auto-Squash-Merge). Gates run independent adversarial checks: claude subagents plus `codex exec` where applicable (different model family, fresh context, no pane setup). Phase 7 auto-squashes bullet commits onto base, deletes the feature branch, removes the worktree and per-worktree Cargo target, and pings DONE-MERGED so sequential chained runs always start from a clean base. Includes the orchestration script, the tmux-pair-orchestration skill, PROJECT.md care, scoped subagents for the gates, sender identity prefixes, explicit parallel-plan markers, auto-detection of repo-specific subagents (`.claude/agents/<repo>-*.md`) listed in every briefing, shipped guidance templates for 7 stacks, bundled companion skills (gepa for prompt-optimization, dg for adversarial code review), and opt-in pi as a bulk-work coding agent. `/run` defaults to codex and switches to claude only for recon-heavy/plan-integration/greenfield/compliance profiles. Multi-pane spawn modes were removed in 0.19.0.

3 on GitHubv0.22.9Apache-2.0updated 5 months agoworkflow
/

Plugin tmux-pair from marketplace derveloper-skills in derveloper/derveloper-skills. Commands read from the repo's manifest — 1 month ago.

What it installs

Skills (3)

  • dg

    Adversarial code review mit Dinesh vs Gilfoyle (Silicon Valley). Aufruf: /dg (git diff), /dg 3 (3 Runden), /dg src/foo.rs (Datei), /dg src/foo.rs 3 (Datei + Runden). Zwei Agenten debattieren ueber Code-Qualitaet: Gilfoyle greift an, Dinesh verteidigt. Bestaetigte Issues werden im Final Summary gesam

  • gepa

    This skill should be used when the user asks about GEPA, prompt optimization, or text artifact evolution. Typical triggers: "optimize a prompt", "improve a system prompt", "tune a prompt with GEPA", "run GEPA optimization", "evolve a prompt", "optimize anything with LLMs", "create a GEPA evaluator",

  • tmux-pair-orchestration

    Use when the user wants to run, monitor, or recover a tmux-pair solo coding workflow through `/run` or `/solo`: one agent in a fresh git worktree, gated recon, plan checks, implementation, final verify, commit hygiene, and auto-squash-merge. Covers durable standards, review-ledger evidence, CLARIFY-

Slash commands (2)

  • /tmux-pair:run

    Auto-entry to tmux-pair. Inspects the repo + task and invokes solo with the right flags. Solo is the only mode.

  • /tmux-pair:solo

    Spawn a single agent in a fresh git worktree, gated by a 7-phase self-driven workflow (recon, plan-check, impl, self-review, PROJECT.md/skill-persist, commit, auto-squash-merge) with subagent-driven adversarial gates

Subagents (5)

  • gate-2-plan-check

    Adversarial plan-check before implementation. Validates that the solo agent's plan covers the task goal-backward, has falsifiable done-definitions per bullet, names files+functions+lines, has tests anchored, explicit parallel markers per bullet, and the 6 frontend-smoke positions for UI bullets. Ret

  • gate-3-code-reviewer

    Adversarial code-reviewer of the diff before final-merge. Checks bugs, security, code quality, anti-AI-slop. Returns VERDICT=PASS|WARNING|BLOCKER with file:line + problem + fix-direction. Read-only. Spawned by tmux-pair solo at Phase 4 in parallel with gate-3-verifier and the codex diff-review secon

  • gate-3-verifier

    Goal-backward verifier after implementation. Reads the plan + the diff (git diff base..HEAD), TRUSTS the TESTS-PROOF marker left in each bullet commit instead of re-running gates, and checks plan-bullet coverage. Runs the project's narrowest test command ONLY when the marker is missing or stale; nev

  • reviewer-readiness-check

    Pre-implementation gate for solo. The solo agent spawns this subagent after GATE 1 (clarify) and BEFORE planning. The subagent reads .claude/rules/*.md, .claude/skills/*/SKILL.md, CLAUDE.md, and the worktree, then scores an 8-item hard checklist (style, tests, architecture, anti-patterns, naming, se

  • rules-bootstrap

    Generates project-specific reviewer guidance from the plugin's language templates plus repo recon plus user answers. Default output is `.claude/skills/<topic>/SKILL.md` (path-scoped, on-demand) so the spawn-time context floor stays small. `.claude/rules/<topic>.md` (always-on) is reserved for truly

View source on GitHub

Similar plugins

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

Backend API design, GraphQL architecture, workflow orchestration with Temporal, and test-driven backend development

9 skills · 1 command · 8 agents

wshobson/agents 42,461

beads

claudecodex

AI-supervised issue tracker for coding workflows

1 skill · 1 agent · hooks

gastownhall/beads 28,698

MiniMax AI skills library for frontend, fullstack, Android, iOS, shader, GIF sticker, document, presentation, spreadsheet, and multimodal media workflows

18 skills

MiniMax-AI/skills 13,828

worktrunk

claudecodex

Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows. This plugin provides configuration guidance (LLM commit messages, project hooks, worktree paths), automatic activity tracking (🤖/💬 indicators in `wt list` showing active Claude sessions), and a `/wt-diff` pane showing each worktree's changes.

2 skills · hooks

max-sixty/worktrunk 12,481

gsd-core

claude

GSD Core is a meta-prompting, context engineering, and spec-driven development system for AI coding agents.

72 skills · 72 commands · 64 agents · hooks

open-gsd/gsd-core 11,494

Maintain tmux-pair?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted