By gethamster
Execute Hamster Studio briefs from Claude Code: schedule tasks in parallel waves, run diff-aware QA, perform paranoid code review, and generate engineering retrospectives.
Implements all subtasks of a single parent Hamster Studio task (HAM-XXX). Reads the parent and all its subtask files from .hamster/, loads project context (project skills, blueprints, methods, CLAUDE.md), discovers relevant codebase context just-in-time, implements all subtasks sequentially in one session, updates task statuses, and reports all changes. Execution-only with leeway: tasks are pre-generated upstream and trusted by default, but stale references are adapted (and documented), and genuine plan defects are escalated as PLAN_ISSUE rather than blindly implemented. Does NOT run project validation — that is handled by the orchestrator after all parallel executors complete. Examples: <example> Context: The orchestrator needs a parent task and its subtasks implemented. assistant: "I'll launch the task-executor to implement HAM-100 and its subtasks HAM-101, HAM-102, HAM-103." <commentary> Use task-executor for each parent task in the execution loop. One agent session handles all subtasks. </commentary> </example>
Reviews and simplifies the cumulative code changes of one execution wave (one or more parent tasks). Phase 1 reviews the full wave diff for convention compliance, quality, security, and completeness — producing a per-parent PASS or NEEDS_FIXES verdict. Because it sees the whole wave, it also catches cross-parent integration issues that per-task review would miss. For parents that pass, Phase 2 applies surgical simplification while preserving all functionality. Runs once per wave, after all parallel task-executors complete and validation/tests pass. Examples: <example> Context: Wave 2 (HAM-100 and HAM-300) finished executing and validation passed. assistant: "Launching wave-reviewer to review the cumulative wave diff and simplify what passes." <commentary> One wave-reviewer per wave reviews all parents together, replacing N per-parent review agents. </commentary> </example>
Plan a Hamster Studio brief. Read tasks, build dependency graph, detect parallel execution waves, with optional CEO or Eng review modes. Use when the user wants to analyze a brief before executing.
Systematic QA with diff-aware, full suite, quick smoke, or regression testing modes. Issue taxonomy and optional fix loop. Use when the user wants to run tests or check quality on their changes.
Resume an interrupted brief execution. Auto-detect progress, reconstruct state from git history and task statuses, and continue from the correct wave. Use when the user wants to continue a previously interrupted ship session.
Engineering retrospective from git history. Team metrics, contributor deep-dives, trends, and actionable insights. Use when the user wants a retrospective on recent development activity.
Paranoid two-pass code review. CRITICAL issues that block shipping, then INFORMATIONAL advisory findings. Interactive resolution for critical issues. Use when the user wants a code review on their current feature branch.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Sync project context from Hamster Studio to your local repositories. Briefs, tasks, blueprints, and methods land in a .hamster/ directory, ready for your editor and AI tools.
curl -fsSL https://tryhamster.com/cli/install | bash
Or download a binary directly from the latest release.
| OS | Architecture |
|---|---|
| macOS | Intel (amd64), Apple Silicon (arm64) |
| Linux | amd64, arm64 |
| Windows | amd64 |
# Authenticate with Hamster Studio
hamster auth login
# Initialize a repository
hamster init
# Sync project context
hamster sync
# Or watch for real-time updates
hamster sync --watch
| Command | Description |
|---|---|
hamster auth login | Authenticate via browser (OAuth 2.1 + PKCE) |
hamster auth logout | Log out and clear stored credentials |
hamster init | Initialize .hamster/ directory and run first sync |
hamster sync | One-time sync from Hamster Studio |
hamster sync --watch | Continuous real-time sync via WebSocket |
hamster status | Show sync status and statistics |
hamster task status <id> <status> | Update task status (todo, in_progress, done) |
hamster brief status <slug> <status> | Update brief status |
hamster tui | Launch Mission Control terminal dashboard |
.hamster/
briefs/ # Project briefs
tasks/ # Task summaries and notes
blueprints/ # Architecture documents
methods/ # Team conventions
A Claude skill is also generated at .claude/skills/hamster-project-context/ for automatic project context awareness.
This repo also ships a Claude Code plugin that orchestrates end-to-end execution of Hamster Studio briefs. The plugin is execution-only: plans (parent tasks, subtasks, context) are generated upstream in Hamster Studio and synced into .hamster/ via hamster sync. The plugin schedules those existing tasks into parallel waves inline (no planner agent), dispatches independent parent tasks simultaneously, reviews each wave, and creates bisectable commits per parent task. Executors load project context as they go — the hamster-project-context skill, project skills, blueprints, and methods — but never generate or elaborate tasks. Trust comes with leeway, not blindness: executors adapt to mechanical drift (a file moved, a helper renamed) and document it, and escalate genuine plan defects as PLAN_ISSUE — verified by the orchestrator, decided by the user when scope is affected, and fed back to Hamster Studio via the PR's Plan Feedback section.
In Claude Code:
/plugin marketplace add gethamster/cli
/plugin install hamster@gethamster-cli
This repository also includes first-class plugin manifests for:
.cursor-plugin/plugin.json and .cursor-plugin/marketplace.json.codex-plugin/plugin.json and .codex-plugin/marketplace.jsonThese manifests include listing metadata (version, author, homepage, repository, license, keywords, category, tags) plus component paths (agents, commands, skills) and a shared logo (assets/logo.svg).
Cursor publish URL:
# Submit for review/public listing
# https://cursor.com/marketplace/publish
Codex note:
.codex-plugin/plugin.json in your Codex host/runtime integration.Codex local install helper:
./scripts/install-codex-plugin.sh # symlink mode (default)
./scripts/install-codex-plugin.sh copy # copy mode
| Skill | Persona | Description |
|---|---|---|
/hamster:ship [slug-or-url] | Release Engineer | Ship a brief: merge base, implement in parallel, test, review, bisectable commits, PR |
/hamster:plan [slug-or-url] | Tech Lead + CEO/Eng modes | Analyze brief with optional founder or architecture review |
/hamster:resume [slug] | — | Resume interrupted execution from where you left off |
/hamster:review | Staff Engineer | Paranoid two-pass code review (CRITICAL then INFORMATIONAL) |
/hamster:qa [mode] | QA Lead | Systematic testing: diff-aware, full, quick, regression |
/hamster:retro [days] | Eng Manager | Engineering retrospective with metrics, trends, team analysis |
/hamster:shipThe main orchestrator. Accepts a brief slug, UUID, or Hamster Studio URL:
/hamster:ship user-authentication
/hamster:ship https://tryhamster.com/home/hamster/briefs/2de8d546-50ab-4dbd-a678-579ec8119f60
npx claudepluginhub gethamster/cli --plugin hamsterTools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
AI-powered development tools for code review, research, design, and workflow automation.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.