Jtsylve Claude Experiments
jtsylve-claude-experiments is a Claude Code plugin marketplace for the Build phase that offers state machine–based prompt optimization and routing plugins for agent workflows.
by jtsylve · github.com/jtsylve/claude-experiments
Browse and install Claude Code optimization plugins that route prompts through state machines, templates, and hybrid agent preprocessing.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install jtsylve-claude-experiments@jtsylve/claude-experimentsBuilt 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:jtsylve/claude-experiments") and the paste-ready config with skillselion.get_install_config("plugin:jtsylve/claude-experiments").
What it does
jtsylve-claude-experiments is a curated Claude Code marketplace for optimization-oriented plugins, centered on state machine–based prompt optimization rather than one-off integrations. Solo builders who treat Claude Code as a product surface use it when default prompting feels noisy, when they want deterministic routing before creative work, or when they need specialized preprocessing templates without maintaining a private plugin repo. The listing emphasizes agents, classification, deterministic and hybrid flows, meta optimization, preprocessing, prompt routing, specialized handlers, state machines, and templates—signals that this is multi-phase agent infrastructure usable during build setup, ship-time review consistency, and operate-time iteration on agent behavior. It is intermediate complexity because you must understand which optimization plugin fits your workflow and how state transitions affect outputs. Confidence is moderate-high given clear marketplace positioning but limited SKILL.md detail in the ingest payload. This is a plugin marketplace for agent ergonomics, not a hosted LLM API or a single crash-ingest tool.
Highlights
- Curated Claude Code plugin marketplace focused on optimization, not random community dumps
- State machine–based prompt optimization with classification and routing keywords
- Hybrid deterministic + specialized agent preprocessing and template flows
- Meta layer spanning agents, meta, preprocessing, and routing—not a single-task MCP shim
- Single marketplace entry (pluginCount: 1) advertising multiple optimization patterns
Why builders use it
Default Claude Code prompting wastes tokens and yields inconsistent routing when you need deterministic preprocessing and specialized agent templates.
After adding the marketplace, you can install optimization plugins that classify tasks, run state-machine prompt flows, and stabilize agent behavior across sessions.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 0 stars, 0 votes.
FAQ
Who is jtsylve-claude-experiments for?
Claude Code users who want curated prompt-optimization and state-machine routing plugins instead of hand-rolling every preprocessing step.
When should I use jtsylve-claude-experiments?
Use it during Build and later Ship/Operate passes when you need classification, templates, and hybrid deterministic routing around your agent prompts.
How do I add jtsylve-claude-experiments to my agent?
Add the jtsylve/claude-experiments Claude Code marketplace source, browse the bundled optimization plugins, and enable the state-machine or routing plugin that matches your workflow.
Comments
Share how you use jtsylve-claude-experiments, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.