agentleFS
harlan-zwharlan-zw/brundlefly/skills/glossary/SKILL.md

glossary

Skill7 starsChanged 33 days agoMITView on GitHub

Creates, extends, or audits GLOSSARY.md, the record of what each product concept is called. Use whenever someone wants one name for a concept, adds a term, wants synonyms such as job, task, and run to stop drifting, or checks naming across the UI, API, and docs.

19 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use recap, test-driven-development and 1 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • genomewalkerSkill

    recap

    Rebuild working context from transcripts and chitta memories in approximately 1500 tokens. Use when starting a new session to continue previous work.

    Also in 3 projects that use this

  • OpenAISkill

    test-driven-development

    Use when implementing any feature or bugfix, before writing implementation code

    Also in 3 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 4 projects that use this

The file

Show its rules and text

We split this skill into the rules it gives an agent

15 rules. 0 are common, 14 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 16 edited versions we read keep these.

  • 14

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Consider their knowledge, context, and reading needs, including ADHD and dyslexia.

    line 10
  2. Distinctive:

    Use one established term for each concept.

    line 14
  3. Distinctive:

    Read the target project's instructions before editing.

    line 18
  4. Distinctive:

    Create or audit GLOSSARY.md only.

    line 20
  5. Distinctive:

    Treat a new product term with the same caution as a public export.

    line 26
  6. Very specific:

    Read GLOSSARY.md before naming anything user-visible: If the repo has one, its terms win over anything that reads better in the moment.

    line 30
  7. Distinctive:

    Never introduce a synonym for a term that exists: If the glossary says Sprint, do not write "run", "batch", or "job", not even in a …

    line 31
  8. Distinctive:

    Keep map labels, table cells, and definitions easy to scan.

    line 45
  9. Distinctive:

    Do not rename public surfaces or remove required maps to make the glossary shorter.

    line 47
  10. Distinctive:

    Before removing a substantial section or feature, explain the proposed loss and ask unless already authorized.

    line 48
  11. Distinctive:

    Read the workflow reference and the format reference before creating or auditing a glossary.

    line 52
  12. Distinctive:

    Use init for a missing glossary, audit for an existing one, and add for an approved new term.

    line 53
  13. Distinctive:

    Ask only about unresolved naming decisions.

    line 55
  14. Distinctive:

    If a decision needs approval, prepare the evidence and proposed terms before asking.

    line 56
  15. Distinctive:

    Keep unresolved choices in Open questions.

    line 57

“Common” is measured: kept by at least 6 in 10 of the 16 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Glossary

Help readers understand product concepts and distinguish them without learning competing names. Consider their knowledge, context, and reading needs, including ADHD and dyslexia. Respect their attention through familiar terms, accurate definitions, and concise explanations that keep necessary distinctions.

GLOSSARY.md owns product terms across UI strings, public APIs, headings, routes, errors, and commit subjects. Use one established term for each concept. Consistency alone does not prove the reader understands it.

Ownership and dependencies

Read the target project's instructions before editing. Respect its branch, worktree, publication, and approval rules. This Skill does not own delivery. Create or audit GLOSSARY.md only. Do not create COPY.md or VISION.md automatically.

The failure mode this exists to stop

Independent naming choices can give one concept different names across the UI, SDK, docs, and routes. Readers must then infer which names mean the same thing. Published names also carry a cost to change. Treat a new product term with the same caution as a public export.

Rules

  1. Read GLOSSARY.md before naming anything user-visible. If the repo has one, its terms win over anything that reads better in the moment.
  2. Never introduce a synonym for a term that exists. If the glossary says Sprint, do not write "run", "batch", or "job", not even in a tooltip, a variable name, or a log line.
  3. Never use a term on the ban list. The ban list carries a replacement; use it.
  4. A concept with no term does not get named silently. Propose an addition, state the candidate term and the synonyms it displaces, and get confirmation. Inventing quietly is the whole failure mode.
  5. Take the platform's word before inventing one. GitHub, Nuxt, Vue, and HTTP have already named most things. auto merge beats a coined merge tier, because the reader knows it and nobody has to confirm it. Propose at most one new term per change; a set of new terms is a redesign, not a name.
  6. Match the recorded casing exactly. Nuxt SEO and NuxtSEO are different brands to a reader.
  7. A term list without a relationship map is half a glossary. See below. Terms are only ambiguous in relation to each other, so the map is what makes the list decidable.

Rule 4 protects naming decisions that nobody has approved yet.

Review for understanding

Identify who uses each term and which distinctions they need to understand. Explain unfamiliar concepts in plain words. Show their relationships and actual use without circular definitions or new synonyms. Apply clarity, including ADHD and dyslexia guidance, to definitions and relationship explanations. Keep map labels, table cells, and definitions easy to scan. Retain complete relationships and exact public identifiers. Preserve established names and casing. If a name confuses readers, show the problem and propose a change for approval. Do not rename public surfaces or remove required maps to make the glossary shorter. Before removing a substantial section or feature, explain the proposed loss and ask unless already authorized.

Workflow

Read the workflow reference and the format reference before creating or auditing a glossary. Use init for a missing glossary, audit for an existing one, and add for an approved new term. Apply the relationship map, frozen-surface checks, decision records, and scope rules in the workflow reference. Ask only about unresolved naming decisions. Existing explicit approval remains valid. If a decision needs approval, prepare the evidence and proposed terms before asking. Keep unresolved choices in Open questions. Do not silently make them canonical.

Scope

Glossary governs nouns for product concepts: what a thing is called. It does not govern voice, tone, or sentence style. A supplied COPY.md owns those. Read it directly. When they disagree on a product noun, the glossary wins; on the sentence around it, COPY.md wins.

A term list found inside a COPY.md belongs here, not there. Step 0 of init already greps for one; fold it in and leave a pointer behind.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.