claude

sa

Simple Agents for Claude Code — command-free skills + subagents for fast single-feature development: plan, approve, isolate in a worktree, implement, PR — Sonnet build and review, Opus only for risky surfaces

3 on GitHubby ken2403v0.1.5MITupdated 1 month agoworkflow
/

Plugin sa from marketplace agent-parallel-dev-plugin in ken2403/agent-parallel-dev-plugin. Commands read from the repo's manifest — 1 month ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (7)

  • apply-feedback

    Addresses review feedback on a PR and updates it, parallelizing across files with implementer subagents when feedback spans 3+ independent files. Use after /sa:review-pr requests changes, or to act on human review comments and push fixes back.

  • clean-worktrees

    Reclaims worktrees after features land — removes every merged worktree in the repo (sa/ha/ca and any other, regardless of location) and deletes their merged branches. Use once a feature PR is merged. Safe by default — never removes a worktree with uncommitted changes, never deletes an unmerged branc

  • code-review

    Standards for writing and reviewing code — quality, test rigor, security, and codebase consistency. Use whenever implementing, changing, or reviewing code, or when checking a PR or diff for bugs, missing test coverage, vulnerabilities, missed propagation, or style drift.

  • merge-pr

    Merge a reviewed sa PR after confirming it is genuinely ready — open, not draft, no changes requested, green CI, mergeable. Use once /sa:review-pr (or a human review) approves the PR and you want it merged from the CLI. Refuses to merge on red checks, changes-requested reviews, or conflicts. Invoke

  • resolve-conflicts

    Resolves merge conflicts between a feature branch and its base branch, in an isolated worktree, parallelizing across files with implementer subagents when 3+ files conflict. Use when a PR or feature branch is behind/conflicting with the base branch and you want the conflicts resolved, verified, and

  • review-pr

    Critically reviews a PR for correctness, security, and codebase consistency — an independent second opinion, not a rubber stamp. Use to review a simple feature's PR before merging. Posts inline comments with --comment.

  • simple-implement

    Takes a plan — a file path or plain-text task — and drives ONE simple feature to an open PR fast, with a human approval gate before any code is written. Use for a small-to-medium, well-scoped implementation task where you want to move quickly and review separately afterward. Stops at PR; run /sa:rev

Subagents (3)

  • deep-verifier

    Opus-tier adversarial reviewer for the claims that decide whether a change ships — the security lens on a risky-surface diff (same wave as the verifiers), the settler of an UNCERTAIN would-be-blocking claim or conflicting verifiers, and the resolve-conflicts integration check. Dispatch with a scoped

  • implementer

    Implements one file-disjoint slice of a simple feature inside an absolute worktree the caller provides. Dispatch several in parallel — one per slice — when a task spans independent files. Slices must be file-disjoint so parallel implementers never collide.

  • verifier

    Adversarial, read-only reviewer that tries to REFUTE a specific claim about a change (correct, safe, complete). The cheap fan-out cross-checker — dispatch several in parallel, one lens each and blind to each other. Use for the pre-PR check, PR review lenses, and feedback re-checks. Defaults to skept

Hooks (1)

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/guard-protected.sh"

View source on GitHub

More from ken2403/agent-parallel-dev-plugin

All 3 plugins →

ha

claudecodex

Claude Code Higher Agents — build ONE feature thoroughly with a deep red-teamed plan, SDD per-task loop, risk-scaled pre-PR gate, independent review, feedback handling, gated merge, conflict resolution, and cleanup. Requires superpowers; model-agnostic

9 skills · 2 agents · hooks

3

ca

claudecodex

Cooperate Agents — Claude×Codex loop: red-team a milestone plan, implement test-first on a draft PR with Claude checkpoints, run a blind-Claude plus bounded offline single-agent Codex final review, promote only on a validated verdict, then merge and clean up worktrees

10 skills

3

Similar plugins

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

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

15 skills · hooks

obra/superpowers 316,836

Use Alibaba Open Code Review from local Codex.

2 skills

alibaba/open-code-review 50,068

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

9 skills · 1 command · 8 agents

wshobson/agents 42,461

Slash commands for quick Repomix operations. Pack local and remote repositories with simple commands like /pack-local and /pack-remote.

2 commands

yamadashy/repomix 29,255

beads

claudecodex

AI-supervised issue tracker for coding workflows

1 skill · 1 agent · hooks

gastownhall/beads 28,698

Maintain sa?

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 plugins3 plugins