Release Readiness: A Visible Skill Map for Agent Shipping

Where this fits

You have updated a skill folder, added a new tool call, and are about to cut a release. The change looks fine in isolation, but you need to prove it meets your release policy and has a rollback owner. release-readiness formalizes that review by turning the folder into a visible Skill Map: it reads the policy, checks the rollback runbook, and gives agents a go/no-go decision instead of a vague 'looks good.'

Why agents benefit

• An agent can read references/release-policy.md and apply hard rules before answering, so a release never gets approved without a cited gate.
• It verifies rollback ownership from references/rollback-runbook.md, blocking risky deployments that lack an owner, command, or monitoring signal.
• The preflight check runs before any review work: if references/ is missing, the agent fails fast instead of hallucinating a verdict.
• The workflow nodes make the agent's reasoning auditable: evidence, policy check, rollback check, then a final decision.
• It turns a free-form review into a structured Skill Map that humans can inspect and reuse across many skill releases.

Practical scenarios

Independent skill audit before tagging

A maintainer asks the agent to review a skill that changed several user-facing messages. The agent collects release facts, compares them against references/release-policy.md, and reports which release gates exist and which are missing.

Rollback ownership for a hotfix

An agent wants to ship a hotfix to a skill used in production. release-readiness checks references/rollback-runbook.md, finds no owner, and returns a blocking no-go with the missing rollback fields. The human adds an owner and reruns the review.

Onboarding a community skill

You download a skill folder from another repository and need to decide whether to adopt it. The review reads the folder's own references if present, or flags that release policy and rollback docs are missing, so you can make an informed adoption decision.

Add it to your agent workflow

Install the skill with:

npx skills add lizhiyao/oh-my-knowledge --skill release-readiness

Then in an agent conversation:

Run release-review on skills/my-skill and tell me if it is safe to ship

The agent will invoke the workflow, read the policy and rollback files in that folder, and return a go/no-go decision with the missing items.

Compare before adopting

Before adding release-readiness, compare it with alternatives on four axes: quality signals, maintenance freshness, alternatives, and workflow fit. Check whether the repository's references are actually maintained, whether the skill's hard rules match your release gates, and whether it supports your folder layout. The README expects references/ or a skill-specific path; if your project uses another structure, you will need to adjust it.

Why it is worth tracking

release-readiness is small and focused, which makes it easy to evaluate. The repository has 18 stars and an active author, but the real signal is whether the policy and rollback rules stay current with how you ship. It is worth tracking now because release approval is a common agent workflow, and this skill makes that workflow visible instead of trusting a model's memory. For a live view of quality and update activity, visit https://www.openagentskill.com/skills/lizhiyao-release-readiness.