DE
deployment-planner
Plan deployment environments, infrastructure, secrets, migrations, workers, schedulers, storage, domains, SSL, and release flow.
Install
mkdir -p .claude/skills/deployment-planner && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/14765" && unzip -o skill.zip -d .claude/skills/deployment-planner && rm skill.zipInstalls to .claude/skills/deployment-planner
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Plan deployment environments, infrastructure, secrets, migrations, workers, schedulers, storage, domains, SSL, and release flow.128 charsno explicit “when” trigger
About this skill
deployment-planner
Token Budget
Medium. Chunk by environment if needed.
When to Use
- Before first deploy.
- Before production release.
- When deploy target/environment changes.
Entry Conditions
- The user request matches this skill.
- Required prior outputs are available or explicitly missing.
Procedure
- Identify deploy target and environments.
- Define secrets/env vars and protection rules.
- Plan build artifacts, migrations, queues/workers, scheduler, storage, cache, domain/SSL.
- Include preflight and smoke checks.
- Hand off rollback requirements to
rollback-planner.
Exit Conditions
- Deploy plan is actionable and rollback requirements are explicit.
Loop Conditions
- If environment conflicts with architecture, route to
architecture-review. - If rollback is incomplete, route to
rollback-planner.
Escalation Conditions
- Escalate to human when requirements conflict, data loss is possible, production risk appears, or safe verification is not possible.
Required Output
- Deployment target.
- Environment map.
- Secrets/env requirements.
- Build/release steps.
- Migration/worker/storage considerations.
- Smoke checks.
- Rollback requirements.
Universal Contract
For every run, report:
- What was inspected.
- What changed, if anything.
- What assumptions were made.
- What verification was run.
- What loop, blocker, or escalation remains.
- Whether
.agent/shared state orcontext/should be updated.
Do not edit files unless the user explicitly requested implementation or this skill's task requires file creation. Do not store secrets in shared state or context.