agentleFS
srikanth-manivannansrikanth-manivannan/claude-agent-factory/skills/architecture/migration-planner/SKILL.md

migration-planner

Skill1 starsChanged 5 months agoMITView on GitHub

Use this to plan a safe, phased migration with rollback using the strangler-fig pattern.

5 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 3

    Distinctive

    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
  1. Distinctive:

    Choose a strategy: Default to strangler-fig (incrementally route traffic from old to new behind a facade).

    line 51
  2. Very specific:

    Plan data migration: Prefer expand/contract (dual-write/backfill/cut-read) for zero-downtime.

    line 56
  3. Distinctive:

    Define rollback per phase: Every phase must be reversible.

    line 59
  4. Distinctive:

    Every phase is reversible: No phase ships without a stated, testable rollback (step 6).

    line 73

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Migration Planner

Use this to plan a safe, phased migration with rollback using the strangler-fig pattern.

Tech profile — Technology: any · Language: n/a · Stack: n/a · Toolchain: n/a · Domain: architecture (Pure-reasoning skill — plans migrations between any two technologies/architectures.)

When to use this skill

  • Migrating from a legacy system/component to a new one with minimal risk.
  • Replatforming, re-architecting, or splitting a monolith incrementally.
  • Any change too large/risky to do as a single cutover.

When NOT to use this skill

  • Choosing what to migrate to — use architecture/tradeoff-analyzer first.
  • A single reversible schema change — use the database-migration workflow.
  • Reviewing a target design — use architecture/system-design-reviewer.

Prerequisites

  • The source (current state) and the target (decided end state).
  • Risk tolerance, downtime budget, and any hard deadline.

Inputs

InputRequiredDescription
sourceyesCurrent system/architecture being migrated from
targetyesDecided end state being migrated to
constraintsnoDowntime budget, deadline, team, compliance
risk_tolerancenoHow much risk is acceptable (defaults to low)

Instructions

  1. Confirm source, target, and "why". Restate both states and the driver. If the target isn't decided, stop and route to architecture/tradeoff-analyzer.
  2. Inventory & assess. List the components/data/integrations in scope; rank by risk and coupling. Identify the seams where old and new can coexist.
  3. Choose a strategy. Default to strangler-fig (incrementally route traffic from old to new behind a facade). Use big-bang only if explicitly justified and low-risk; record the rationale.
  4. Define phases. Break the migration into independently shippable, independently reversible phases. Each phase has: scope, entry/exit criteria, and verification.
  5. Plan data migration. Prefer expand/contract (dual-write/backfill/cut-read) for zero-downtime; specify integrity checks. (For the mechanics, the database-migration workflow applies.)
  6. Define rollback per phase. Every phase must be reversible; state the rollback trigger and steps. A phase without a tested rollback is not ready.
  7. Sequence & gate. Order phases by risk (de-risk early); define the go/no-go gate between phases.
  8. List risks + assumptions. Top risks with mitigations; explicit assumptions.

Output

A phased migration plan: source/target/driver, scope inventory, chosen strategy (+why), ordered phases (scope, criteria, verification, rollback), the data-migration approach, a risk register with mitigations, and assumptions.

Constraints & safety

  • Every phase is reversible. No phase ships without a stated, testable rollback (step 6). Aligns with the deployment-review checklist.
  • De-risk early. Sequence the riskiest, most-uncertain phases first when feasible.
  • No silent data loss. Data phases specify integrity verification.
  • Conforms to /standards/prompt-engineering.md and /standards/security.md.

Examples

Minimal below; full plans in EXAMPLES.md.

Given: source = monolith auth, target = dedicated auth service, low risk tolerance. Produces: a strangler-fig plan — facade/routing phase → read-path migration → write-path migration → decommission, each with rollback and verification.

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.