Agent skill
new-command
Scaffold a new CLI command for Cortex TMS following the existing codebase patterns.
Install this agent skill to your Project
npx add-skill https://github.com/cortex-tms/cortex-tms/tree/main/.claude/skills/new-command
SKILL.md
New Command
Scaffold a new CLI command for Cortex TMS.
Before Writing Code
- Read existing patterns — look at src/commands/archive.ts and src/commands/status.ts as reference implementations
- Read types — check src/types/cli.ts for existing interfaces
- Read tests — check src/tests/archive.test.ts for the test pattern
- Propose the command — describe what it does, its subcommands/options, and which files will be created/modified. Follow Propose/Justify/Recommend. Wait for approval.
Implementation Checklist
After the user approves the plan:
- Create
src/commands/<name>.ts— exportcreateXCommand()andxCommand - Add types to
src/types/cli.tsif needed - Register in
src/cli.ts— import andprogram.addCommand() - Create
src/__tests__/<name>.test.ts— follow existing test patterns - Add section to
README.mdunder CLI Commands - Update
NEXT-TASKS.md— check off the task
Rules
- Match the exact patterns in existing commands (Commander.js, chalk, ora)
- Follow existing test patterns (vitest, createTempDir, runCommand)
- If the user asks a question, STOP and answer before continuing
- Show git diff before committing. Wait for approval.
Arguments
$ARGUMENTS — the name and purpose of the new command
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
release
Guide through the Cortex TMS release process — version bump, changelog, sync, tag, publish. Every step requires user approval.
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?