ea-service
Skill3 starsChanged 3 months agoMITView on GitHub
Operate the architecture-service layer (Architecture-as-a-Service / on-demand) - maintain the service catalog, intake and fulfil service requests, keep SLA hygiene. Use when someone requests something from EA ("we need a review / context / an exception"), when defining or changing EA's offerings, or when REQ006-REQ008 findings (SLA breaches, retired offerings) need working.
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
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
- Very specific: line 54
Fulfil: Run the offering's fulfilment path.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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
The architecture service layer
EA here is a provider, not a gate: a catalog of offerings with owners and SLAs
(services/, one per file), and a demand ledger of who asked for what
(governance-log/requests/). Consumption is recorded, so EA's value is measurable
instead of asserted -- and maintenance follows demand instead of ambition.
The catalog
# services/svc-context-pack.yaml
id: svc-context-pack
name: Agent context pack
description: What the consumer gets, phrased from the consumer's side.
fulfilledBy: ea-context / easkills context --scope <element>
owner: ea@example.test
slaDays: 2
lifecycle: active # proposed | active | retired
selfService: trueCatalog discipline: an offering is a promise with a number -- no owner or SLA, no
offering (schema rejects it). Keep the catalog short and true: five offerings people
use beat fifteen that decorate. Retire honestly (lifecycle: retired); requests
against retired offerings are flagged (REQ008). selfService: true marks offerings
the consumer can run themselves -- prefer growing that set; every self-service
fulfilment is an SLA that cannot be breached.
The request lifecycle
# governance-log/requests/req-2026-08-erp-context.yaml
id: req-2026-08-erp-context
service: svc-context-pack
requestedBy: erp-replatforming-team@example.test
requested: 2026-08-01
scope: [app-erp-core]
status: open # open | fulfilled | declined- Intake. Record the request the moment it arrives, with the offering id and
the model elements in
scope(they must exist --REQ004; if the area is unmodelled, that is intake work scoped to this request, on-demand, thin-slice: never model beyond what the request needs). - Fulfil. Run the offering's fulfilment path; set
status: fulfilled,fulfilled:date andfulfilledBy:pointing at the deliverable (REQ005rejects unevidenced fulfilment). The deliverable carries its own "as of" freshness -- never strip it. - Or decline, with a reason in
notes(REQ007). Declining out-of-scope work honestly is part of being a service; silent queues are not. - SLA hygiene. An open request past its offering's
slaDayswarns (REQ006) and shows inkpias a breach. The honest responses: fulfil, decline with reason, or renegotiate the catalog promise -- never quietly edit dates.
What demand buys you
easkills kpi-- the Service line: offerings, open/fulfilled/declined, SLA breaches, average fulfilment time. This is the AaaS value evidence.easkills staleness-- every element carries its demand count; review what people ask about first, and treat never-requested content as a candidate for de-scoping rather than eternal upkeep.
Reporting back
The request's disposition (fulfilled with what, or declined why), SLA state against the promise, and anything the request revealed: an unmodelled area, a missing offering people keep asking for informally, or an offering nobody has requested in months -- the catalog is maintained by the same evidence discipline as the model.
In this repository25 other files it gives its agents.
- ea-adrSkill
- ea-ai-governanceSkill
- ea-alignSkill
- ea-approveSkill
- ea-boardSkill
- ea-capability-mapSkill
- ea-change-triageSkill
- ea-checkSkill
- ea-complianceSkill
- ea-contextSkill
- ea-delta-ingestSkill
- ea-dispensationSkill
- ea-docsSkill
- ea-evalSkill
- ea-healthSkill
- ea-importSkill
- ea-intakeSkill
- ea-modelSkill
- ea-regulatorySkill
- ea-runSkill
- ea-securitySkill
- ea-stakeholdersSkill
- ea-standards-baseSkill
- ea-validateSkill
- ea-viewsSkill
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.

