seed-skills/release-readiness-checklist/SKILL.md
Teach agents to build go or no-go release readiness scorecards with gated criteria for coverage, flakes, defects, performance, security, and sign-off.
npx skillsauth add PramodDutta/qaskills Release Readiness ChecklistInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
You are a release quality lead who builds a clear go or no-go scorecard using objective gates, risk notes, owner sign-off, and evidence from tests, defects, performance, security, and operations.
Create a small release readiness workspace.
mkdir -p release/readiness release/reports release/sql
touch release/readiness/scorecard-template.md
touch release/sql/release-quality.sql
Define release inputs.
// release/readiness/types.ts
export type GateStatus = 'pass' | 'warn' | 'fail' | 'not_applicable';
export type ReleaseGate = {
name: string;
status: GateStatus;
threshold: string;
evidenceUrl: string;
owner: string;
notes: string;
};
export type ReleaseScorecard = {
releaseName: string;
buildId: string;
decision: 'go' | 'no_go' | 'pending';
gates: ReleaseGate[];
};
Use this scorecard for every candidate.
# Release Readiness Scorecard
Release:
Build:
Date:
Decision: Pending
Release owner:
QA owner:
Engineering owner:
## Gates
| Gate | Status | Threshold | Evidence | Owner | Notes |
|---|---|---|---|---|---|
| Unit and integration tests | Pending | 100 percent pass | CI | QA | |
| E2E smoke | Pending | 100 percent pass | CI | QA | |
| Open S1 defects | Pending | 0 | Jira | Product | |
| Open S2 defects | Pending | Approved only | Jira | Product | |
| Flake budget | Pending | Under 2 percent | Test dashboard | QA | |
| Performance budget | Pending | No critical regression | Report | Web lead | |
| Security gate | Pending | No high or critical | Scan | Security | |
| Rollback plan | Pending | Verified | Runbook | DevOps | |
Automate the score where data is available.
// release/readiness/evaluate.ts
import type { ReleaseGate, ReleaseScorecard } from './types';
function hasFailedGate(gates: ReleaseGate[]): boolean {
return gates.some((gate) => gate.status === 'fail');
}
function hasPendingGate(gates: ReleaseGate[]): boolean {
return gates.some((gate) => gate.status === 'not_applicable' ? false : gate.status === 'warn');
}
export function evaluateRelease(releaseName: string, buildId: string, gates: ReleaseGate[]): ReleaseScorecard {
const decision = hasFailedGate(gates) ? 'no_go' : hasPendingGate(gates) ? 'pending' : 'go';
return { releaseName, buildId, decision, gates };
}
Use SQL when release evidence is stored in a warehouse.
select
suite_name,
count(*) as total_tests,
sum(case when status = 'passed' then 1 else 0 end) as passed_tests,
sum(case when status = 'failed' then 1 else 0 end) as failed_tests,
round(100.0 * sum(case when status = 'passed' then 1 else 0 end) / count(*), 2) as pass_rate
from test_results
where release_name = '2026.07'
group by suite_name
order by suite_name;
Use explicit criteria for the meeting.
Record sign-off as a decision log.
## Sign-Off
QA: Approved, smoke and regression gates pass.
Engineering: Approved, rollback plan verified.
Product: Approved, accepted S2 risk QA-4281.
Security: Approved, no high or critical findings.
DevOps: Approved, monitors active and release window staffed.
Decision: Go
Time:
Decision owner:
| Gate | Pass | Warn | Fail | |---|---|---|---| | Open S1 defects | 0 | Not used | Any open S1 | | Open S2 defects | 0 | Accepted risk | Unapproved S2 | | Smoke tests | 100 percent pass | Retry under review | Any confirmed fail | | Flake budget | Under 2 percent | 2 to 5 percent | Over 5 percent | | Performance | Within budget | Small accepted regression | Critical path regression | | Security | No high or critical | Medium findings tracked | High or critical |
development
Generate test data from the schemas you already have. Read OpenAPI, JSON Schema, SQL DDL, or TypeScript models and produce deterministic factories, boundary and negative cases, relational datasets with valid foreign keys, cleanup scripts, and PII-safe synthetic data. Production records never leave the machine.
development
Diagnose flaky test failures from Playwright reports, traces, and rerun history. Classify each failure as product, test, environment, data, or unknown with cited evidence and a proposed fix. Never auto-modifies code without opt-in.
tools
Test LLM, RAG, MCP, and agentic systems end to end. Build golden datasets, run deterministic checks and LLM judges, score retrieval, probe prompt injection, verify tool use, and gate CI on thresholds. Orchestrates DeepEval, Ragas, promptfoo, and Langfuse.
development
Analyze a git diff, map affected risks, select the tests that matter, detect coverage gaps on changed lines, run configurable quality gates, and produce a go/no-go release report with cited evidence. Recommends only; never merges or deploys.