Agent skill
release
Guide through the Cortex TMS release process — version bump, changelog, sync, tag, publish. Every step requires user approval.
Install this agent skill to your Project
npx add-skill https://github.com/cortex-tms/cortex-tms/tree/main/.claude/skills/release
SKILL.md
Release
Guide the user through a Cortex TMS release.
Pre-release Gate
Before starting, run the full quality gate:
pnpm test— must passpnpm run lint— must have 0 errorspnpm run build— must succeednode bin/cortex-tms.js validate --strict— must pass
If any step fails, STOP and report. Do not proceed with a broken release.
Release Steps (each requires user approval)
-
Bump version — show the new version number and the command. Wait for approval.
bashnpm version [patch|minor|major] -
Sync version tags — update all version markers across docs.
bashnode scripts/sync-project.js -
Update CHANGELOG.md — draft release notes from commits since last tag. Show draft, wait for approval.
-
Commit release — show commit message, wait for approval.
-
Create git tag — show tag command, wait for approval.
bashgit tag v[version] -
Push — show push command, wait for approval.
bashgit push origin main --tags -
Publish to NPM — show publish command, wait for approval.
bashnpm publish -
Create GitHub release — draft release notes, wait for approval.
Rules
- NEVER skip a step
- NEVER proceed without explicit user approval at each gate
- If something fails mid-release, stop and explain how to recover
Arguments
$ARGUMENTS
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
new-command
Scaffold a new CLI command for Cortex TMS following the existing codebase patterns.
sync
Synchronize task files and source-of-truth documents. Update NEXT-TASKS.md, FUTURE-ENHANCEMENTS.md, README, and CHANGELOG to reflect current project state.
implement
Implement an approved plan. Write code, tests, and docs following the sprint checklist in NEXT-TASKS.md.
plan
Research the codebase and create an implementation plan following Propose/Justify/Recommend. Use this before any feature, fix, or refactor.
validate
Run the full quality gate — tests, lint, build, and cortex-tms validate --strict.
verl-rl-training
Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
Didn't find tool you were looking for?