Coalesce Labs Catalyst
coalesce-labs-catalyst is a Claude Code plugin for the Build phase that provides a token-efficient workspace with parallel agents, memory, and a Research → Plan → Implement → Validate workflow.
by coalesce-labs · github.com/coalesce-labs/catalyst
Run a token-efficient Claude Code workspace that chains research, planning, implementation, and validation with parallel agents and memory.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install coalesce-labs-catalyst@coalesce-labs/catalystBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:coalesce-labs/catalyst") and the paste-ready config with skillselion.get_install_config("plugin:coalesce-labs/catalyst").
What it does
coalesce-labs-catalyst is a Claude Code plugin collection pitched as a token-efficient workspace for solo builders who want discipline without enterprise overhead. It packages five plugins that support a repeatable Research → Plan → Implement → Validate cycle, leaning on parallel agents and persistent memory so context survives across sessions instead of being re-explained every prompt. The catalog metadata highlights product management rhythms—backlog grooming, milestones, standups—and optional bridges to GitHub, Linear, PostHog, and Sentry, which makes the pack relevant when you are both shipping features and watching funnels or incidents. Because the workflow is methodological rather than a single integration, you can invoke Catalyst thinking during validation scoping, deep implementation, and pre-ship validation alike. Builders who already juggle Claude Code for coding and separate tools for PM may install Catalyst to keep research notes, plans, and implementation traces in one agent-native system with lighter token burn than unstructured multi-thread chatting.
Highlights
- Documented Research → Plan → Implement → Validate workflow for Claude Code
- Token-efficient workspace design with parallel agents and persistent memory
- Five plugins spanning PM, analytics (PostHog), debugging (Sentry), and dev integrations
- Optional GitHub–Linear stack for grooming, milestones, and handoffs
- Community-sourced patterns for discovery, incidents, and production monitoring hooks
Why builders use it
Ad-hoc Claude Code sessions waste tokens and lose context, so research, planning, and shipping feel like disconnected chats.
You get a structured agent workspace with five plugins that carry memory forward and walk features from research through validation.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 9 stars, 0 votes.
FAQ
Who is coalesce-labs-catalyst for?
Indie developers and small teams who want Claude Code to orchestrate research, planning, implementation, and validation with less token churn.
When should I use coalesce-labs-catalyst?
Use it when starting or refactoring a project cycle and you want parallel agents plus persistent memory across the four workflow stages.
How do I add coalesce-labs-catalyst to my agent?
Install the coalesce-labs/catalyst marketplace plugin in Claude Code, enable the five bundled plugins, and configure optional GitHub, Linear, PostHog, or Sentry integrations you rely on.
Comments
Share how you use coalesce-labs-catalyst, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.