mintuz/skills

Reusable development skills for Codex, Claude and OMP. The new workflow and principle skills use portable Markdown instructions; host-specific discovery and tools remain the harness’s responsibility.
## Install with Codex
bash
codex plugin marketplace add mintuz/skills
Restart the ChatGPT desktop app, open the Plugins Directory, choose
Mintuz Skills, and install the bundles you need.
## Install with Claude Code
bash
# Add the marketplace
/plugin marketplace add https://github.com/mintuz/skills
# Install plugins
/plugin install core@mintuz-skills
/plugin install web@mintuz-skills
/plugin install typescript@mintuz-skills
/plugin install system-design@mintuz-skills
/plugin install product-management@mintuz-skills
/plugin install app@mintuz-skills
/plugin install life@mintuz-skills
/plugin install principle@mintuz-skills
## Use with OMP or standalone skill loaders
Keep each complete skill directory, including references. For the 54 skills adapted from pstack (7 core workflows and 47 principles), the copier adds THIRDPARTYNOTICES.md so the MIT notice travels with the bundle. To generate flat directories with the existing copier:
bash
go run scripts/codex-sync.go --output /path/to/your/skills
Point the harness’s skill discovery at that directory, or copy it into its documented skill location. The copier preserves ordinary SKILL.md bundles; despite its name it does not translate the instructions into a Codex-only format. Leave --prefix off to preserve names such as principle-prove-it-works.
Use one authoritative distribution per skill name. Run omp --help for the installed version’s discovery/filter options. Portable prose alone is not a runtime certification.
## When to use the main skills
This section covers the entry points and the skills that I use every day. The Skills tables list every other skill. The example prompts come from real Claude Code and OMP sessions. Project names and links are replaced with placeholders.
Each harness calls a skill by name with a different syntax. The examples use the Claude Code form.
| Harness | Syntax | Example |
| Claude Code | /name | /reducer |
| OMP | /skill:name | /skill:reducer |
| Codex | $name | $reducer |
### Pick a skill
| Situation | Skill |
| A task needs a full workflow: review, bug fix, spec, implementation, design | adam-mode |
| An open "which approach?" question needs an experiment before the real build | adam-mode Prototype (spike) playbook |
| One artifact must reach a high, confirmed bar over several rounds | gauntlet-loop |
| One artifact can take several shapes, and you want them judged in one pass | arena |
| The look of a UI concept is open, and the variants must differ | creative-seed |
| Existing code or a system has more moving parts than its behaviour needs | reducer |
| New work must stay at the minimum that the requirement needs | yagni |
| You need to know what the code does now | how |
| You need to know why the code exists in its current form | why |
### adam-mode
adam-mode is the main entry point. It is manual-only, so call it by name. It matches the request to a playbook, loads the skills that the playbook needs, delegates to adam-agent and proves the result before it reports. It continues without asking on reversible work, such as branches, commits, PRs and CI repair. It stops before a merge, the auto-merge label, a force-push, a data deletion or a message to a person, unless you asked for that action in the session.