Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
Ville Laitila / Softagram Oy avatar

Softagram Plugins

  • Updated June 1, 2026
  • softagram/claude-plugins-marketplace

softagram-plugins is a Claude Code marketplace that distributes GoodReason, a four-role agent topology so developers can tackle complex engineering tasks with separated strategy, design, implementation, and verification.

About

softagram-plugins is a Claude Code marketplace distributing Softagram's GoodReason multi-agent plugin. Instead of one agent doing everything, it cycles Strategist, Architect, Implementer, and Evolution subagents with strict role separation and a coordinator. Reach for it on complex engineering or diagnostic tasks where a single agent's blind spots—self-review, hypothesis testing, and causal verification—need structured separation.

  • Four subagents: Strategist, Architect, Implementer, Evolution
  • Systems-thinking meta-ontology with strict role boundaries
  • TDD phasing and phi-stuck probe-test protocol when blocked
  • Coordinator routing via /goodreason:cycle slash command

Softagram Plugins by the numbers

  • Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add softagram/claude-plugins-marketplace

Add your badge

Show developers this marketplace is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Last updatedJune 1, 2026
Repositorysoftagram/claude-plugins-marketplace

How do I stop one AI agent from both writing code and judging its own fixes on complex or diagnostic tasks?

Runs GoodReason's four-role agent cycle—Strategist, Architect, Implementer, Evolution—for complex engineering tasks.

Who is it for?

Engineers facing ambiguous bugs or multi-step features who want role-separated subagents with TDD and verification gates.

Skip if: Simple one-file edits or teams that prefer a single undifferentiated agent without orchestration overhead.

What you get

A paced multi-agent cycle producing hypotheses, structure, tested implementation, and Evolution-verified outcomes with causal checks.

Recommended Marketplaces

FAQ

What agents does GoodReason use?

Strategist for foundations and hypotheses, Architect for structure, Implementer for TDD code, and Evolution for verification and feedback.

How do I start a cycle?

Install goodreason@Softagram Plugins, then run /goodreason:cycle with your task description in Claude Code.

Where are full definitions documented?

Agent definitions live in the goodreason-agents-setup repository; GOODREASON.md covers the complete meta-ontology reference.

AI Agentsagents

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.