Unique-Divine/jiyuu / ai-skills/lexicon-domain-lang
lexicon-domain-lang
VerifiedThe lexicon skill. Keep a project's domain language in LEXICON.md files. Use when the user asks for the lexicon skill, asks what to call something or which word is right, wants terminology defined or cleaned up, or mentions a lexicon, glossary, domain language, ubiquitous language, or LEXICON.md. Covers which lexicons apply to a path, using their terms, challenging conflicting words, and recording a term once it is settled.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.8 KB |
| THIRD_PARTY.md | markdown | 1.2 KB |
| references/format.md | markdown | 3.2 KB |
references/format.md
View on GitHub# LEXICON.md Format
## Template
```md
# Lexicon: Ordering
Ordering receives customer orders and tracks them until delivery.
- Parent: [Shop](../LEXICON.md)
- Children: [Returns](returns/LEXICON.md)
- Related: [Billing](../billing/LEXICON.md): shares **Customer**; Billing
invoices the Orders that Ordering marks delivered.
## Terms
**Order**:
A customer's request to buy one or more items.
_Avoid_: purchase, transaction
**Delivered**:
The carrier confirmed handover to the customer.
_Avoid_: shipped, completed
_See_: [delivery states](docs/delivery.md)
## Never Say
- "account" for a person: say **Customer** or **User**; they differ.
```
## Parts
- **Title and description.** `# Lexicon: <Context Name>`, then one or two
sentences on what this part of the project is.
- **Links.** Up to three lines, each present only when it has a target:
- `Parent`: the nearest lexicon in an ancestor directory, however many
levels up.
- `Children`: lexicons in subdirectories.
- `Related`: lexicons of other contexts, each with the shared term and one
line on how the two contexts relate.
- **Terms.** One entry per concept:
- `**Term**:` then one or two sentences saying what the thing is.
- `_Avoid_:` the other words people use for it. Optional.
- `_See_:` links to the specs, ADRs, or docs that explain or decide it,
each with a few words on what it holds. Optional.
- `_Overrides_:` a link to the parent lexicon, when this context gives a
parent's term a different meaning. Optional and rare.
- **Never Say.** Words ruled out that have no single replacement, each with
what to say and, where it helps, why.
## Rules
- **One word per idea.** Pick the best word; list the rest under `_Avoid_`.
- **Define what it is, not how it works.** One or two sentences.
- **Project terms only.** A general programming word belongs only if the
project gives it a specific meaning.
- **No implementation detail.** No line numbers, signatures, or algorithms.
Put those behind a `_See_` link.
- **Group when clusters appear.** Use `###` subheadings under `## Terms` once
a file passes about ten terms. A short file stays flat.
- **Sort for the reader.** Order terms so each builds on the ones above it;
alphabetical order is a fallback.
- **Keep it short.** A lexicon is read at the start of many tasks. Remove a
term that no longer causes confusion and has no synonym to rule out.
## Scope and Links
- A `LEXICON.md` applies to its directory and below. A child inherits every
ancestor term and repeats none, except a term it marks `_Overrides_`.
- The same word may mean different things in two contexts. State the
difference in `Related` on both sides.
- Inside a repository, link with relative paths. Across repositories, link
with the hosting URL so the link works in any checkout.
- A public repository's lexicon links only to public material.
## Where Files Go
- One at the repository root when the project has one shared language.
- One in a subdirectory when that part has words of its own, such as a tool
in a monorepo or a module with a distinct domain.
- Create each lazily, when its first term is settled. One exception: a file
may exist with links and no terms when it is the parent that ties child
lexicons together.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |