← Claude plugins
CLAUDE PLUGIN

mash-multi-agent-software-harness

Publisher not specified · Claude Code

What this plugin does

MASH (Multi-Agent Software Harness) is a structured development framework for Claude Code that manages the full software lifecycle — from idea to tested, committed code. It orchestrates four specialized AI personas (Init, Plan, Dev, QA) that work in sequence with strict boundaries: planning personas interact with you to capture intent, while implementation personas run as isolated sub-agents that write code and tests autonomously.

Features are tracked through a clear lifecycle (CREATED → DEV_READY → WIP → DONE), with automatic retry logic (up to 3 attempts) and failure analysis fed back into subsequent attempts. All specs live in .mash/plan/ as the single source of truth — immutable during development.

Key features:
- Interactive planning that asks clarifying questions before writing specs
- Isolated dev and QA sub-agents with strict read/write boundaries
- Git workflow options: per-feature worktrees or current branch
- Auto-commit after QA pass, or manual git control
- Feature retry with failure context propagation
- Self-updating via /mash update

Catalog observations

Sources
Community
Works with
Claude Code
Reported installs
Not provided
First observed
2026-10-03 22:13 UTC
Last observed
2026-10-11 00:00 UTC

Installation counts are reported by Claude Marketplace, not independently verified active users. Observation dates are not release dates. Pricing and account requirements must be checked with the publisher.

Install in Claude Code

Choose a source and review what the plugin adds before installing.

Community

/plugin marketplace add anthropics/claude-plugins-community
/plugin install mash-multi-agent-software-harness@claude-community

Claude Code installation documentation

Source evidence

Listing and manifest JSON

Descriptions and compatibility labels come from the linked sources. Where both sources match, the GitHub description is used because web summaries may be shortened. We have not independently tested the declared capabilities.