deployment-readiness

STANDARD OPERATING PROCEDURE

Purpose

Assess whether a service is ready to ship by validating requirements, risks, rollback posture, and observability hooks.

Trigger Conditions

• Positive: Pre-flight for production deployment; Go/no-go decision needed; Release checklist review
• Negative: Active incident recovery (route to cicd-intelligent-recovery); Marketing or launch orchestration (route to sop-product-launch); Post-release ops maturity audit (route to production-readiness)

Guardrails

• Structure-first: keep SKILL.md aligned with examples/, tests/, and any resources/references so downstream agents always have scaffolding.
• Adversarial validation is mandatory: cover boundary cases, failure paths, and rollback drills before declaring the SOP complete.
• Prompt hygiene: separate hard vs. soft vs. inferred constraints and confirm inferred constraints before acting.
• Explicit confidence ceilings: format as 'Confidence: X.XX (ceiling: TYPE Y.YY)' and never exceed the ceiling for the claim type.
• MCP traceability: tag sessions WHO=operations-{name}-{session_id}, WHY=skill-execution, and capture evidence links in outputs.
• Avoid anti-patterns: undocumented changes, missing rollback paths, skipped tests, or unbounded automation without approvals.

Required Artifacts

• SKILL.md (this SOP)
• metadata.json for registry
• Add examples/tests/resources if missing during execution

Execution Phases

1. Clarify deployment scope

   • Identify change set, owners, and environments
   • List hard/soft/inferred constraints and confirm risky areas
   • Check dependencies and service level objectives

2. Evaluate readiness signals

   • Review test coverage, CI status, and security checks
   • Verify observability, feature flags, and config toggles
   • Ensure data/backfill/migration plans are reversible

3. Risk and rollback planning

   • Define rollback/roll-forward criteria and time limits
   • Pre-stage smoke tests and monitors
   • Assign decision owners and approval gates

4. Decide and document

   • Publish go/no-go decision with evidence
   • Capture gaps, mitigations, and follow-up tasks
   • Update runbooks and notify stakeholders

Output Format

• Deployment readiness brief with scope and risks
• Go/no-go criteria with owners and time boxes
• Rollback/roll-forward plan and validation steps
• Evidence links for tests, security, and observability
• Follow-up actions and runbook updates

Validation Checklist

• Constraints confirmed and ambiguous items resolved or flagged
• Rollback path validated or rehearsed
• Critical tests and monitors in place with evidence links
• Decision recorded with approvals and communications
• Confidence ceiling stated for readiness call

Confidence: 0.70 (ceiling: inference 0.70) - structured pre-flight checklist with explicit rollback planning