Flora131 Atomic
flora131-atomic is a Claude Code plugin for the Build phase that helps you compose an agent harness to research codebases, plan complex features, and ship them through iterative autonomous loops.
by flora131 · github.com/flora131/atomic
Install when you want a reusable agent harness to research unfamiliar repos, plan multi-step features, and drive iterative autonomous implementation loops in Claude Code.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install flora131-atomic@flora131/atomicBuilt 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:flora131/atomic") and the paste-ready config with skillselion.get_install_config("plugin:flora131/atomic").
What it does
flora131-atomic is a community Claude Code plugin bundle aimed at solo and indie builders who treat the coding agent as a sustained collaborator, not a one-shot autocomplete. The positioning—build your own harness, understand any codebase, plan complex features, and ship them autonomously—maps to a multi-phase workflow: discovery and scoping on unfamiliar trees, structured planning for non-trivial work, then repeated iterative execution until outcomes land. Keywords such as continuous development, interactive loops, Ralph, Wiggum, and self-referential technique signal methodology over a narrow MCP connector. Use it when you are onboarding to a legacy or third-party repo, when a feature spans many files and needs a plan before edits, or when you want the agent to keep working through a checklist with explicit loop discipline. It complements task-specific skills; it does not replace domain linters, test runners, or production monitoring. Confidence is moderate where SKILL.md-level invoke triggers are not inlined in the catalog ingest—tags follow README positioning and plugin keywords.
Highlights
- Build-your-own harness workflow for Claude Code instead of ad-hoc prompts per repo
- Codebase understanding and research pass before planning complex features
- Feature planning artifacts that feed autonomous shipping loops
- Iterative, self-referential development loops (Ralph/Wiggum-style continuous implementation)
- Community plugin bundle with 1 packaged plugin from flora131/atomic (~136 GitHub stars)
Why builders use it
Jumping straight into edits on unfamiliar or large codebases wastes tokens, misses dependencies, and leaves complex features half-implemented without a repeatable agent workflow.
You get a structured harness mental model—research, plan, then loop on implementation—so Claude Code can drive multi-step features with clearer scope and less manual re-prompting.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 136 stars, 0 votes.
FAQ
Who is flora131-atomic for?
Indie and solo developers using Claude Code who want continuous, iterative agent development across research, planning, and shipping—not casual one-off questions.
When should I use flora131-atomic?
Use it when onboarding to a new codebase, scoping a large feature, or running self-referential implementation loops until the feature is actually shipped.
How do I add flora131-atomic to my agent?
Register the flora131/atomic community plugin in Claude Code from the repo bundle (1 plugin), then invoke its skill(s) from your plugin marketplace or local plugin path per your Claude Code setup.
Comments
Share how you use flora131-atomic, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.