concept-review
Skill4 starsChanged 6 months agoMITView on GitHub
Use when reviewing naming, vocabulary drift, invented concepts, misleading abstractions, or unclear domain language.
In this repository30 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .cursor/rules/agent-feed.mdcCursor rule
- design-reviewSkill
- docxSkill
- engineering-planningSkill
- frontend-designSkill
- guidance-promoterSkill
- project-architectureSkill
- project-developmentSkill
- project-fixSkill
- project-reviewSkill
- skill-creatorSkill
- skill-maintainerSkill
- specialist-routerSkill
- xlsxSkill
- concept-reviewSkill
- design-reviewSkill
- docxSkill
- engineering-planningSkill
- frontend-designSkill
- guidance-promoterSkill
- project-architectureSkill
- project-developmentSkill
- project-fixSkill
- project-reviewSkill
- skill-creatorSkill
- skill-maintainerSkill
- specialist-routerSkill
- xlsxSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
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
- Distinctive: line 10
Use this skill to keep AI-assisted work from introducing unnecessary concepts, misleading names, or vocabulary drift.
- Distinctive: line 14
Use this skill when:
- Distinctive: line 39
If existing terminology is already inconsistent, do not use the inconsistency as permission to add another term.
- Distinctive: line 57
Prefer the plainest accurate name that a future maintainer can understand without translating it back to the implementation.
- Very specific: line 61
Do not rename stable public, persistence, API, CLI, or domain contracts without applying .agents/rules/decision-gates.md.
- Distinctive: line 62
Do not block useful domain terms just because they are new.
- Distinctive: line 63
Do not treat style preference as a finding unless it creates ambiguity, drift, or maintenance cost.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Concept Review Skill
Use this skill to keep AI-assisted work from introducing unnecessary concepts, misleading names, or vocabulary drift.
Required Use
Use this skill when:
- A review includes naming, terminology, abstraction, module boundaries, or user-facing concept changes.
- A diff introduces new nouns, categories, layers, services, records, statuses, protocols, or workflow names.
- The user questions whether wording, naming, or abstraction is necessary.
- A task imports, creates, or modifies external/custom skills whose terminology could override project language.
Required Reading
.agents/rules/outcome-boundary.md.agents/skills/README.md.agents/project/README.md.agents/domain/README.md- The files, diff, plan, or skill being reviewed.
Evidence Priority
Resolve terminology against evidence in this order:
- Current user request and Task Brief language.
- Project/domain source-of-truth files and indexed read triggers.
- Existing public contracts, CLI/API names, generated file names, schemas, and module owners.
- Established naming patterns in the same repository.
- External or custom skill terminology, only as a lower-priority method vocabulary.
If existing terminology is already inconsistent, do not use the inconsistency as permission to add another term. Prefer convergence toward the clearest existing source of truth, or apply .agents/rules/decision-gates.md when renaming a stable contract would affect future development.
Concept Noise Categories
Flag these four types of concept noise:
Vocabulary drift: a new word competes with an existing word for the same concept.Pseudo-concept: a new class, layer, workflow name, status, or category wraps an existing action without adding durable meaning.Semantic misdirection: a name implies a lifecycle, orchestration, state machine, authority, or guarantee that the implementation does not provide.Concept pollution: a lower-level module, reusable rule, public contract, or generic skill imports terms that belong to a different layer, one repository, or one product domain.
Review Procedure
- Identify the baseline vocabulary from project/domain source-of-truth files.
- List new or changed terms in the current work.
- For each term, check whether it maps to an existing project concept, contract, or user-visible behavior.
- Classify any problem as vocabulary drift, pseudo-concept, semantic misdirection, or concept pollution.
- Flag invented concepts that hide a simpler action, duplicate existing language, imply unsupported behavior, or move project-specific meaning into reusable rules.
- Prefer the plainest accurate name that a future maintainer can understand without translating it back to the implementation.
Guardrails
- Do not rename stable public, persistence, API, CLI, or domain contracts without applying
.agents/rules/decision-gates.md. - Do not block useful domain terms just because they are new; require evidence that the term owns a durable behavior or boundary.
- Do not treat style preference as a finding unless it creates ambiguity, drift, or maintenance cost.
Output
Return concept findings with the affected term, why it is risky, the source-of-truth it conflicts with or lacks, and the smallest safer wording or boundary.
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
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.

