claude

rgh-mms

Development control-plane agent surface for Russell's Mini-me Swarm: planner/implementor/reviewer/approver/researcher agents, delivery/review/research skills, and the canonical policy/roles/workflows they follow, self-contained with a layered per-repo config resolver.

0 on GitHubby Russell Heillingv0.1.1Apache-2.0updated 19 days agoworkflow
/

Plugin rgh-mms from marketplace rgh-plugins in chewcorp/rgh-market. Commands read from the repo's manifest — 19 days ago.

What it installs

Skills (9)

  • contract-first-change

    Drives a load-bearing change through contract scaffold, implement, review, and settle. Use when manifest contracts.enabled is true and the change affects harness or runtime behavior.

  • document-review

    Supervises documentation lens checks on doc-only diffs — style, specificity, and prompt-engineer gate for instruction files. Use for policy, README, AGENTS.md, and harness prose changes without behavior edits.

  • documentation

    Docs-only change loop for a registered bundle — implementor then reviewer, lighter than full milestone delivery. Use for README, AGENTS.md, and guide updates without behavior changes.

  • implement-feature

    End-to-end feature implementation for a pinned product repo — plan, implement, review, approve, PR, then land per delivery.integration_mode (human per-issue or milestone-aggregate). Use for issues and milestones.

  • investigate-failure

    Read-only investigation of a failure — CI break, regression, incident. Returns structured findings and recommended fix path without editing code unless scope expands.

  • milestone-delivery

    Supervises a milestone or parent-issue container for a registered bundle through to completion, one issue at a time, via plan -> implement -> review -> approve -> PR -> land (human per-issue or agent-to-milestone then human bulk). Use when asked to deliver, drive, or continue a milestone or parent i

  • research

    Read-only exploration for a registered bundle. Returns a structured research brief without code changes. Use for spikes, design questions, or planner open_questions.

  • review-change

    Review a branch or PR against harness policy and product conventions. Dispatches reviewer role only — no implementation.

  • update-agent-instructions

    Docs-only update to AGENTS.md, local-policy, or harness-generated adapters. Uses documentation workflow with accuracy gates.

Subagents (5)

  • approver

    Verifies acceptance criteria mechanically with evidence per criterion. After plan-time AC rewrite/waiver, honors work-item annotations and waived defect-id Input; no re-fail of the same defect.

  • implementor

    Implements a plan or review findings on an issue branch. Runs local CI, commits, returns branch/commit/summary and ci_evidence.

  • planner

    Planner for one work item. Returns blockers, acceptance_criteria_defects, ordered subtasks, files touched, and acceptance criteria. Never writes code or git; local-mirror populate may create tracker sub-issues per ${CLAUDE_PLUGIN_ROOT}/roles/planner.md.

  • researcher

    Read-only research. Returns structured brief with evidence and recommendation.

  • reviewer

    Read-only review against harness policy and product conventions.

View source on GitHub

More from chewcorp/rgh-market

All 4 plugins →

rgh-sme

claude

Zettelkasten-style LLM wikis across a set of subject-matter domains, built from reference texts and packaged as Open Knowledge Format (OKF) bundles.

25 skills

0

Harness-agnostic curation tooling for Zettelkasten-style OKF wiki bundles: skills, shared agents, and scripts to onboard sources, dispatch domain curators, review charters, and drain scoring/landscape backlogs. Domain-agnostic — reads domain slugs from the consuming repo's okf-core.toml. Mastered directly in this repo (not vendored from Cloudsmith).

7 skills · 4 agents

0

rgh-pre

claude

Consumer wiki package for personal wellbeing and Personal Reliability Engineering: philosophy, mindset, telemetry-state, capacity-load, failure-modes, controls-design, incident-response, reliability-strategy.

9 skills

0

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

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

Maintain rgh-mms?

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 plugins4 plugins