agentleFS
prabhakar349prabhakar349/claude-ddd-redesign/skills/interface-inventory/SKILL.md

interface-inventory

Skill1 starsChanged 3 months agoMITView on GitHub

Defines the content and structure of the Interface Inventory — synchronous endpoints and asynchronous streams, published and consumed. Reference when assembling or reviewing inventory/interfaces.json.

The file

Files
SKILL.md

Interface Inventory

Defines what an Interface record contains. Does not define how to discover interfaces — that is the discovery agents' job (agents/sync-surface-extractor.md, agents/async-surface-extractor.md), reading from the language pack (seam S3.2).

Schema

schemas/interface.schema.json. One JSON object per interface; inventory/interfaces.json is a JSON array of these.

Required fields

id (IFC-SYNC-NNNN or IFC-ASYNC-NNNN), kind (sync | async-published | async-consumed), name, owning_service, evidence (non-empty).

Content obligations by kind

  • sync — populate the sync object: method, path, request/response schema if resolvable, observed status codes.
  • async-published / async-consumed — populate the async object in full: topic, payload schema, schema registry ID if any, partition key, ordering guarantee, delivery semantics, idempotency key, retry policy, DLQ, retention. Endpoint-only inventories are a documented failure of prior art (REQ-01) — async fields are not optional decoration.
  • consumers — every entry needs identifier, evidence_source, confidence. consumers_complete defaults to false for async interfaces and must stay false unless both schema-registry evidence and consumer-group evidence support true. Do not set it true on partial evidence.

Behavior-ID citation

An Interface does not cite Behaviors directly — Behaviors cite Interfaces (behavior.interfaces[]). The traceability spine runs Behavior → Interface, not the reverse; do not add a behaviors field to an Interface record.

Unknowns

Any field you cannot determine is "unknown" where the schema has that enum value, or omitted with the gap noted in evidence — never a guess (Constitution C10).

Minimal worked example

{
  "id": "IFC-ASYNC-0007",
  "kind": "async-published",
  "name": "order.created",
  "owning_service": "orders-svc",
  "protocol": "kafka",
  "async": {
    "topic": "order.created",
    "partition_key": "order_id",
    "ordering_guarantee": "per-partition",
    "delivery": "at-least-once",
    "dlq": "order.created.dlq"
  },
  "consumers": [
    { "identifier": "billing-svc", "evidence_source": "consumer-group", "confidence": "high" }
  ],
  "consumers_complete": false,
  "evidence": [ { "source": "code", "locator": "orders/publisher.py:41" } ]
}

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.