Prompts are text you paste into Claude Code. Drills are executable skills that Claude Code loads from ~/.claude/skills/. Same vendor, completely different purpose. This page exists because buyers ask which to buy first. Short answer: Prompts if you're under $10 / still evaluating, Drills if you want consistent output every time.
Skills that live in ~/.claude/skills/ and enforce scope, output shape, and sequence. Same structure every invocation.
Text prompts you paste into Claude Code directly, or paste into your CLAUDE.md file. Raw inputs; Claude handles the rest.
| Drills | Prompts Pack | |
|---|---|---|
| Price | $29 lifetime | $9 lifetime |
| Item count | 25 skills | 24 prompts |
| How you use it | Installed in ~/.claude/skills/, invoked by name | Copied into chat or CLAUDE.md as raw text |
| Claude Code specific? | Yes — requires the skills directory | No — works with any Claude client (desktop, API, web) |
| Output consistency | Enforced via skill definition | Depends on how you invoke it and what context is loaded |
| Updatability | git pull and new drills load automatically | Manual re-paste when we ship new prompts |
| Install friction | 1 clone + 1 cp command | Zero — just paste |
| Refuses-to-do rules | Built into each skill (non-negotiable) | Not enforced; relies on you not asking for bad behavior |
| Best when... | You use Claude Code daily and want consistent output without re-prompting | You're evaluating, switching tools often, or want the under-$10 entry point |
This is your first Septim purchase. You use multiple Claude clients (not just Code). You want to spend under $10 to test the Septim house style. You don't want to manage a ~/.claude/ directory. You're okay with paste-and-tweak being the invocation model.
You live in Claude Code day-to-day. You're tired of re-writing the same review prompt five different ways and getting five different outputs. You want the output shape enforced (severity + file:line + action). You ship to prod and want review drills that refuse to approve without citing verified findings. The $29 gap over Prompts Pack is paid back the first time a drill catches a bug a loose prompt would have missed.
You run a solo-founder stack and want the full Septim Claude Code kit. The Claude Code Mastery Bundle at $69 adds Agents Pack ($49 alone) to Drills + Prompts — net savings $18 off individual pricing. If you're buying two of the three anyway, the third is effectively free.
Drills + Agents + Prompts together for $69 — save $18 vs separate. Three private repos, lifetime access, one payment.
See the Mastery Bundle — $69 →