# Release

> Pre-release validation for MirrorBuddy via app-release-manager subagent Use when this capability is needed.

- Skill: `tomevault-io/release-77` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add tomevault-io/release-77`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/release-77/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: tomevault-io (https://skillmd.com/u/tomevault-io)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/tomevault-io/release-77

---


# Release Manager - MirrorBuddy

Pre-release validation for MirrorBuddy via `app-release-manager` subagent.

## Context (pre-computed)

```
Project: MirrorBuddy
Branch: `git branch --show-current 2>/dev/null || echo "not a git repo"`
Uncommitted: `git status --short 2>/dev/null | wc -l | tr -d ' '` files
Version: `node -p "require('./package.json').version" 2>/dev/null || echo "unknown"`
```

## Activation

When message contains `/release` or `/release {version}`.

## What It Does

Launches `app-release-manager` subagent with MirrorBuddy-specific validations:

1. **Educational Content Quality** - AI tutors, knowledge bases, teaching prompts
2. **Student Safety** - Content filtering, bias detection, safety guardrails
3. **GDPR Compliance** - Data handling, privacy, consent mechanisms
4. **Accessibility (WCAG 2.1 AA)** - 7 DSA profiles, contrast, keyboard nav
5. **ISE Engineering Fundamentals** - Code quality, testing, documentation
6. **AI Tutor Readiness** - All 26 maestri verified, tools working

## Workflow

### Phase 1: Launch Agent

```typescript
await Task({
  subagent_type: 'app-release-manager',
  description: 'MirrorBuddy release validation',
  prompt: `
RELEASE VALIDATION FOR MIRRORBUDDY

Project: MirrorBuddy
Target Version: ${version || 'auto-detect from package.json'}
Branch: ${branch}

Execute full release gate:
1. Pre-flight checks (git clean, correct branch)
2. Build validation (lint, typecheck, build)
3. Test execution (unit, E2E)
4. Security audit (secrets, dependencies)
5. Accessibility audit (WCAG 2.1 AA)
6. Compliance check (GDPR, EU AI Act)
7. Sentry + deployment configuration verification (CI + Vercel env)
8. AI tutor verification (all 26 maestri)
9. Documentation review (CHANGELOG, ADRs)

Sentry + deployment checks MUST include:
- Running \`npm run sentry:verify\` locally
- Ensuring \`.github/workflows/ci.yml\` \`sentry-config\` job installs dependencies and has \`VERCEL_TOKEN\` wired
- Verifying GitHub Actions secrets and Vercel production env contain all required Sentry and core app variables

Zero tolerance policy:
- ANY test failure = BLOCK
- ANY security issue = BLOCK
- ANY a11y violation = BLOCK
- ANY compliance gap = BLOCK
- ANY Sentry misconfiguration or missing env var = BLOCK

Output: Release report with APPROVE or BLOCK decision.
`,
});
```

### Optional Pre-Flight: GitHub Secrets Sync

Before running the full release gate you MAY want to sync your local `.env` into GitHub Actions secrets (for example after adding `VERCEL_TOKEN`, `SENTRY_DSN`, or other critical config). This is a **manual, explicit step** and should only be used when you are sure your local `.env` reflects the intended production values.

```bash
# Sync all KEY=VALUE pairs from .env to GitHub Actions secrets
# Requires GitHub CLI (gh auth login)
npm run secrets:sync
```

Notes:

- This script never logs secret values (only names and lengths).
- It can overwrite existing secrets with the same keys; use it carefully.
- It is NOT run automatically by the release agent to avoid accidental propagation of local-only values.

### Phase 2: Review Results

Agent returns comprehensive report with MirrorBuddy-specific checks.

### Phase 3: User Decision

If APPROVED → version bump, git tag, CHANGELOG, GitHub release
If BLOCKED → fix issues, re-run `/release`

## MirrorBuddy-Specific Checks

| Check   | Validation                                |
| ------- | ----------------------------------------- |
| Maestri | All 26 load correctly, knowledge embedded |
| Coaches | 6 coaches with valid prompts              |
| Buddies | 6 buddies with age-adaptive prompts       |
| Tools   | All 16 learning tools functional          |
| Tiers   | Trial/Base/Pro limits enforced            |
| Safety  | Bias detector, content filter active      |
| A11y    | 7 DSA profiles, instant access button     |

## Zero Tolerance Policy (Blocking)

- ANY compiler/lint warning
- ANY test failure
- ANY security vulnerability
- ANY TODO/FIXME in code
- ANY hardcoded secrets
- ANY debug prints (console.log)
- ANY outdated deps with CVEs
- ANY a11y violation (WCAG 2.1 AA)
- ANY GDPR/EU AI Act compliance gap

## Quick Reference

```bash
# Fast local gate (PR-like)
npm run release:fast

# Full release validation (pre-production)
npm run release:gate

# Sentry + deployment configuration verification (must be clean)
npm run sentry:verify

# Agent-driven full release validation
/release           # Auto-detect version
/release 1.5.0    # Validate specific version
```

## Related

- Agent: `app-release-manager`
- Hardening: `mirrorbuddy-hardening-checks`
- Thor: `thor-quality-assurance-guardian`
- Gate scripts: `npm run release:fast`, `npm run release:gate`

---
> Converted and distributed by [TomeVault](https://tomevault.io/claim/fightthestroke) — claim your Tome and manage your conversions.
<!-- tomevault:4.0:skill_md:2026-04-11 -->