cutover-plan
Skill1 starsChanged 3 months agoMITView on GitHub
Defines the content and structure of a Context's cutover plan — Stage 8 Transition's shadow/canary/cutover/decommission sequencing. Reference when assembling or reviewing contexts/<id>/cutover-plan.json.
In this repository17 other files it gives its agents.
- acceptance-criteriaSkill
- adjudication-recordSkill
- adrSkill
- behavior-inventorySkill
- characterization-suiteSkill
- context-modelSkill
- contract-testsSkill
- design-docSkill
- divergence-registerSkill
- interface-inventorySkill
- migration-waypointsSkill
- observability-specSkill
- retirement-registerSkill
- roadmapSkill
- slice-outlineSkill
- tactical-modelSkill
- variation-policySkill
The file
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 20
transition_profile: must be set explicitly (seam S1.5) — v1 ships "shadow-canary-cutover-decommission".
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Cutover Plan
Defines what a Context's cutover plan contains. Runs per Modeled Context, only once every Slice in that Context has reached Accepted — you cannot canary one Slice of a Context whose siblings are unbuilt, because the Context is the deployable boundary (07-stages.md §1).
Schema
schemas/cutover-plan.schema.json. A single JSON object per Context — contexts/<id>/cutover-plan.json is not an array.
Required fields
context, transition_profile, steps (≥1).
Content obligations
transition_profilemust be set explicitly (seam S1.5) — v1 ships"shadow-canary-cutover-decommission"; the Stage 8 command selects a profile rather than hard-coding steps, so a future specification-oracle profile (e.g. direct launch, no legacy shadow) can be added without rework.- Steps run in order:
shadow(domain logic executes fully, effecting adapters stubbed, decisions diffed) →canary→progressive-cutover→decommission. Each step records itsdiff_rate_thresholdanddiff_rate_observed— the gate is diff rate below threshold at each step, not just at the end. interfaces_retiredlists every legacy Interface this Context's cutover retires, withzero_traffic_days— the Stage 8 gate requires zero traffic for N days before retirement, per Interface, not per Context in aggregate.- Sequencing across Contexts goes by blast radius — read-only and non-effecting Contexts first, state-mutating and externally-effecting Contexts last (
07-stages.mdStage 8).
Behavior-ID citation
The cutover plan does not cite Behaviors directly — it operates on Slices (already Accepted, already characterized) and Interfaces (already covered by contract tests). Its diffs are behavioral in effect but recorded as diff-reports/*, not as new Behavior citations.
What this skill does not cover
Consumer-facing migration guides and deprecation notices — those belong to the Consumer Migration Workstream (retirement-register skill, /ddd-migrate), which runs in parallel and never gates this plan.
Minimal worked example
{
"context": "CTX-0001",
"transition_profile": "shadow-canary-cutover-decommission",
"steps": [
{ "name": "shadow", "status": "complete", "diff_rate_threshold": 0.01, "diff_rate_observed": 0.002 },
{ "name": "canary", "status": "in-progress", "diff_rate_threshold": 0.005 }
],
"interfaces_retired": [ { "interface_id": "IFC-SYNC-0001", "zero_traffic_days": 0 } ],
"decommission_checklist_ref": "contexts/CTX-0001/decommission-checklist.md"
}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.

