- Home
- Skills
- Technical Documentation
- Domain Language Governance
Works with the AI Agents you already use
Domain Language Governance
Resolve terminology drift across specs, APIs, and code into context-specific canonical terms with definitions, invariants, and aliases, while keeping behavior-changing questions open for a domain decision.
$12
Domain Language Governance
Example session with this skill installed
The API, database, and docs use "user," "account," and "member" interchangeably. Based only on the supplied authority and examples, define canonical terms with meaning, context, invariant, and authority; map deprecated aliases and artifacts to align; and identify open domain questions. Do not infer an unverified product rule or claim files changed.
- Read your context and instructions
- Compiled the domain language governance
Illustrative output based on the synthetic example included with the package. These names and findings are not repository evidence; no project files were inspected or changed.
CANONICAL TERMS
| TERM | MEANING | CONTEXT | INVARIANT | AUTHORITY |
|---|---|---|---|---|
| Account | Billing entity that owns a subscription | billing | Exactly one owner | ADR-009 |
| User | A person who can authenticate | identity | Belongs to at least one Account | ADR-009 |
| Member | Role of a User within an Account | authorization | User and Account pair | New glossary entry |
CONFLICTS / DEPRECATED ALIASES
- The
userstable stores Accounts in this synthetic example; route a concept rename and migration to the implementation phase. member_idrefers to User in some API uses and to a membership row in others; split the ambiguous reference.
ARTIFACTS TO ALIGN
- Glossary
- OpenAPI schema
- User and Account tables
- Authentication middleware
OPEN DOMAIN QUESTIONS
- Can a User belong to multiple Accounts simultaneously? UNVERIFIED. A product decision is needed before changing the schema.
Connects securely to your tools. The creator never sees your data.
About this skill
Domain Language Governance resolves competing names and overloaded terms across project contexts into canonical records grounded in project authority and representative evidence. Use it when the same concept has different names across specs, APIs, code, or agent instructions, or when one word means different things across services. It is not for style-only renaming of local variables.
What it does
- Reads the smallest sufficient glossary, ADR, spec, API, type, test, and implementation evidence, recording each term source and observed meaning.
- Classifies conflicts as synonym duplication, overloaded term, missing concept, wrong abstraction boundary, or stale artifact.
- Defines canonical terms from authoritative product or domain evidence, with context, invariant, allowed aliases, and ambiguous usage to avoid.
- Stops for a domain decision when uncertainty changes behavior or contracts. If only a bounded sample is available, repository-wide terminology coverage remains UNVERIFIED.
- Maps the glossary, ADR, spec, API/schema, tests, or code that may need alignment, without mass-renaming implementation while meaning is unsettled.
Output
A canonical-terms record with each term meaning, context, invariant, and authority, followed by conflicts or deprecated aliases, artifacts to align, and open domain questions.
Boundaries
Project governance documents, accepted ADRs, boundaries, glossaries, and approved specs outrank the skill. It does not replace project-specific rules or guarantee model output. Missing evidence, capability, or authority stays explicit instead of being invented. A style-only local variable rename is outside scope.
The result depends on accessible project authority and representative source evidence. It cannot settle a missing product decision when terminology changes behavior or contracts. When only a bounded sample is available, repository-wide terminology coverage remains UNVERIFIED. It maps artifacts to align but does not claim a mass rename, commit, push, install, or deployment.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 17 days ago
Permissions
Read Files is required for project glossaries, ADRs, specifications, and representative APIs, types, tests, and implementation evidence. Terminal/Shell is optional only for targeted search when available; use host-native or manual search when it is not, and mark repository-wide coverage UNVERIFIED when evidence is bounded. Write Files, Browser, Network Access, MCP, external accounts, and Environment Variables are not required by the package. Commits, pushes, merges, installs, deployments, and implementation changes are not implied.