Skip to content

Latest commit

 

History

History

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 
 
 

README.md

naming

A Claude Code plugin for generating and evaluating name candidates. One skill, one job: when a name was just rejected, or none exists yet, break the anchor by fanning out blind, fresh-context generators from distinct lenses, then hand the human a scored shortlist to choose from.

Skill What it does
/naming:name-it-better Generate fresh name candidates from blind lenses, score them, recommend a shortlist. The human picks

Why blind generators

The dominant trigger is a reactive retry: a name was suggested and rejected, and the same context that produced it will only produce more of the same. So the generators run BLIND to the conversation. Each is seeded with only a structured context brief: responsibility, firing context, scope boundaries, collision vocabulary, and a word-level blocklist. Rejected NAMES stay on the main thread's reject list and never enter the brief. Each generator works a distinct lens: responsibility-literal, moment-of-use, or domain-lore. Diverge widely from independent perspectives, then converge once, and keep the first-seen suggestion from anchoring the choice. The method grounding is in the skill's context/sources.md.

The human always picks. The skill narrows and recommends; it never auto-locks a name.

/naming:name-it-better              # blind lenses → scored shortlist → human picks
/naming:name-it-better tournament   # ~5 generators + elimination rounds for high-stakes names

tournament is an honest adaptation of elimination brackets plus pairwise social-choice scoring, not a documented naming technique. The skill presents it as such.

Consumer conventions

  • Criteria source of truth. The skill scores against the consuming organization's naming and domain-language conventions when the project declares them (its CLAUDE.md / .claude/rules/ or a standards source it points to). When none is declared, it falls back to a research-ordered general criteria priority, semantic accuracy first and trigger utility last, grounded in the skill's context/sources.md. A criterion the user wants that is missing from their conventions flows UP into those conventions, not into the skill.
  • Adjacent capabilities. Resolving what a domain concept IS routes to a domain-modeling capability; propagating a chosen name across call sites routes to a rename-references capability. Each is invoked through its slash command when present, and degrades to prose guidance when absent.

Install

/plugin marketplace add melodic-software/claude-code-plugins
/plugin install naming@melodic-software

Configuration

No userConfig. No persistent state. Network: the generators are subagents; context/sources.md carries external reference links for the method, not fetched at runtime.