codename
Use when a repository, project, service, or workstream needs a name and a descriptive one would be dull — 'what should I call this repo', 'give me a…
Skill by LRNZ09
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/lrnz09-codename
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/codename
What's inside
- Skill name
codename- Description
- Use when a repository, project, service, or workstream needs a name and a descriptive one would be dull — 'what should I call this repo', 'give me a codename', 'name this project', naming a spike, experiment, or internal tool, or replacing a placeholder name. Tuned for repo and internal names, not public brand or trademark work.
- License
- MIT
- Instructions
- 97 lines in SKILL.md · view on GitHub
Files (2)
- GENERATIONS.md
- SKILL.md
SKILL.md (first lines)
# Codename You're a naming consultant, not a name generator. Read what the project *is*, decide which aesthetic world it belongs to, hand back one confident pick. Two axes decide everything: - **Generation** — the aesthetic world the project belongs to. Library: [GENERATIONS.md](GENERATIONS.md). - **Scope** — its role, weight, and character. Table below. The name must be unrelated to the implementation. Generation gives the word its world, scope gives it its weight. Neither describes the code. ## Infer, don't interview Gather context yourself — the prompt, the repo you're standing in (README, manifest, directory names), what the user has been building. Then decide. Ask at most one question, and only when you still can't say what the project *does*. Never ask the user to pick the generation or the scope; that's the job they handed you. If you're about to type "which aesthetic did you have in mind?", you already have enough. Choose, and say why. Reading a brief: *"a small CLI for syncing files, something with that early-2000s optimistic internet feeling"* → generation: Frutiger Aero, scope: tiny utility → **breeze**. With no aesthetic signal in the brief, pick the generation from how the project *behaves*, never from what it's built with. Something that runs unattended at 3am belongs to a different world than something people open with their coffee. ## Scope sets the weight | Scope | The name should feel | |---|---| | tiny utility, experiment | light, one or two syllables — a passing effect or a small object | | library | quiet, self-contained: a substance or instrument other things are made *with* | | CLI, tool | crisp and typeable; it has to look right after a `$` | | backend, service | steady, mid-weight, something that runs continuously | | infrastructure | load-bearing — geological, structural, orbital | | user-facing app | warm and inviting: a place, or a light source | | core, flagship | singular and low-frequency — a word there's only one of | | experimental, weird | oblique, unresolved, a little wrong on purpose | | legacy, compatibility | dated on purpose — a superseded technology, worn with affection | ## Screen every candidate Seven tests. One failure kills the candidate. 1. **Out loud** — a colleague hears it once and spells it right. 2. **Standalone** — it works as a proper noun: "Lightship handles that." No "the … service" scaffolding needed. 3. **Shelf** — it would *not* look at home in a list of frameworks and package managers. 4. **Opacity** — nobody can guess what the code does from the name. 5. **Seam** — one word, or a compound that reads as one. An aesthetic noun bolted to a tech noun (NeonBackend) dies here. 6. **Collision** — for anything public, one search: no well-known product owns it in this space. 7. **Mouth** — say it slowly and hear the words hiding inside it: no body part, bodily function, insult or punishment, in *any* language the owner speaks, not just English. `penates` dies here — an Italian ear gets *pena* (punishment) long before the household gods. ## Output Pick, alternates, reason — in that order, under 100 words total. No narration of the process. Given the brief *"a reconciliation service that runs nightly and nobody thinks about"*, the entire reply is: > **lightship** — signals & lighthouses · steady service > > Alternates: pharos, dogwatch, fathom, nocturne > > A lightship is moored where a lighthouse can't stand: unglamorous, always lit, holding its > position all night. Right weight for something that only gets noticed when it stops. Alternates are 3–5 real options, not filler — no near-synonyms of the pick, and at least one from a neighbouring generation. Add a fourth line, `Fit: …`, only when you're *not* confident, naming the one thing that would sharpen it. ## Keep one universe - If `~/.claude/codenames.md` exists, read it first: never re-suggest a name it lists, and let what's there set the house style. Once the user picks, append `- name — repo — generation`. If that file doesn't exist, leave it alone unless asked. - Two consecutive requests shouldn't land in the same generation — unless the projects are siblings in one family, in which case they should, and the names should rhyme in register. - A project type never maps to a fixed name. Same brief, different day, different word. ## Common mistakes | Mistake | Fix | |---|---| | Asking which vibe the user wants | Infer it. One question at most, only about what the project does. | | Explaining the naming process | Three lines, then stop. | | Five variations on one word | Alternates should disagree with each other. | | Reaching for Nexus, Vantage, Apex, Atlas | Generic-startup smell. Fails the shelf test. | | A name that hints at the code (`syncr`, `queuely`) | Fails opacity. Back to the generation, start over. | | Inventing a word when a real obscure one exists | Real words carry more. Invent only when it's phonetically clean. | | Trusting a classical or foreign word because its meaning fits | Meaning is not the only thing it carries. Run the mouth test out loud first. |
From LRNZ09/skills, MIT.
More from LRNZ09/skills
- 0.570score
record-user-journey
LRNZ09
Use when you need a recording or visual proof of a real user journey through a running app — a demo video for a PR, ticket or stakeholder, evidence that a fix…
SkillAFreshness: AClaude Code +9 - 0.504score
diagnose
LRNZ09
Disciplined diagnosis loop for hard bugs and performance regressions.
SkillBFreshness: BClaude Code +9 - 0.499score
council
LRNZ09
Use when the user wants a decision, idea, or piece of work pressure-tested from multiple independent angles before they commit to it — not for questions with…
SkillBFreshness: BClaude Code +9 - 0.499score
tdd
LRNZ09
Test-driven development with red-green-refactor loop.
SkillBFreshness: BClaude Code +9 - 0.499score
improve-codebase-architecture
LRNZ09
Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/.
SkillBFreshness: BClaude Code +9 - 0.499score
grill
LRNZ09
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree one at a time with a…
SkillBFreshness: BClaude Code +9 - 0.499score
caveman
LRNZ09
Ultra-compressed communication mode.
SkillBFreshness: BClaude Code +9 - 0.488score
handoff
LRNZ09
Compact the current conversation into a handoff document for another agent to pick up.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 29, 2026 (29 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/LRNZ09/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from LRNZ09
- 0.570score
record-user-journey
LRNZ09
Use when you need a recording or visual proof of a real user journey through a running app — a demo video for a PR, ticket or stakeholder, evidence that a fix…
SkillAFreshness: AClaude Code +9 - 0.504score
diagnose
LRNZ09
Disciplined diagnosis loop for hard bugs and performance regressions.
SkillBFreshness: BClaude Code +9 - 0.499score
council
LRNZ09
Use when the user wants a decision, idea, or piece of work pressure-tested from multiple independent angles before they commit to it — not for questions with…
SkillBFreshness: BClaude Code +9 - 0.499score
tdd
LRNZ09
Test-driven development with red-green-refactor loop.
SkillBFreshness: BClaude Code +9 - 0.499score
improve-codebase-architecture
LRNZ09
Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/.
SkillBFreshness: BClaude Code +9 - 0.499score
grill
LRNZ09
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree one at a time with a…
SkillBFreshness: BClaude Code +9
Cite this page
codename skill. Agent Mart. https://agentmart.dev/skills/lrnz09-codename (last updated 2026-10-24)