cwilliams5/pre-release
.claude/skills/pre-release/SKILL.md
Pre-flight checklist — run high-signal review skills before a release
$ install --global
skillsauthnpx skillsauth add cwilliams5/Alt-Tabby pre-releaseInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 24, 2026, 8:59 PM2.1s1 file scanned
SKILL.md
- name:
- pre-release
- description:
- Pre-flight checklist — run high-signal review skills before a release
- user-invocable:
- true
- disable-model-invocation:
- true
Run the curated pre-release review checklist. Each review enters planning mode and produces findings. Stop and report after each tier — the user decides whether to proceed to the next tier or address findings first.
Tier 1 — Correctness (bugs that ship to users)
Run these first. Any findings here are release-blockers:
/review-race-conditions— New races = crashes or data corruption in the wild/review-reentrancy— STA pump reentrancy = shader corruption, visual artifacts, frozen paints (Critical "On" does NOT prevent this)/review-resource-leaks— Leaks degrade the user's system over time/review-option-interaction— Broken install/update/admin path = users stranded
After Tier 1: Report findings. If any exist, ask the user: address now or proceed to Tier 2?
Tier 2 — Hygiene (things that rot if not caught)
/review-debug— Ungated log writes or accidentally-enabled diagnostics ship to users
After Tier 2: Report findings. Ask the user: address now or proceed to release?
Execution
For each skill in order:
- Invoke the skill (it enters planning mode and produces a plan)
- Summarize findings concisely — count of issues by severity
- If zero findings, note it and move to the next skill
- After completing a tier, present the tier summary and pause for user decision
Reporting
After all skills have run, present a release readiness summary:
| Skill | Findings | Severity | |-------|----------|----------| | review-race-conditions | 0 | — | | review-reentrancy | 0 | — | | review-resource-leaks | 2 | 1 high, 1 low | | review-option-interaction | 0 | — | | review-debug | 1 | medium (ungated log) |
Release recommendation: Proceed / Address N issues first
The user makes the final call. This skill surfaces information — it does not block releases.
Related Skills
cwilliams5/worktree
tools
VerifiedTrustedCommunity
Create a new git worktree and switch the session into it
1SKILL.mdUpdated Apr 16, 2026
cwilliams5/trace
tools
VerifiedTrustedCommunity
Spawn agent to trace code flow via query tools — answer only, no context cost
1SKILL.mdUpdated Apr 16, 2026
cwilliams5/ship
tools
VerifiedTrustedCommunity
Commit, push, and create a PR for the current branch
1SKILL.mdUpdated Apr 16, 2026
cwilliams5/shader-retire
tools
VerifiedTrustedCommunity
Retire a shader by moving its files to legacy/shaders_retired
1SKILL.mdUpdated Apr 16, 2026