smithery/hello-pork-belly

hlab-planner

Designs atomic specs from ambiguous requests when you need a modular plan; use it to define scope, architecture, and inputs. Output is a docs/specs/<TASK_ID>-SPEC.md with user story, flow diagram, file manifest, env interface, and step logic.

Installation

$ npx skills add smithery/hello-pork-belly --skill hlab-planner

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery/hello-pork-belly.

npx skills add smithery/hello-pork-belly

Browse all from smithery/hello-pork-belly

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Skill metadata

Parsed from SKILL.md frontmatter.

Version2.0
More metadata
version
2.0
architecture
Wizard-Executor-Separation

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 1,662 B
  • docs SUMMARY.md 262 B

History

  1. First recorded snapshot · 0 installs

SKILL.md

HLab Planner Skill

Core Philosophy

You are designing for the Horizon CLI Framework, a modular Linux ops system. Rule #1: Separation of Concerns.

  • Wizard (Brain): Pure interaction. Collects inputs. Generates a .env file. NEVER modifies the system.
  • Executor (Hand): Pure execution. Reads .env file. Idempotent. Logs to file. NEVER asks questions.

Knowledge Base (Constraints)

  1. LOMP-Lite: Target RAM < 1GB. Must enforce Swap checks, PHP concurrency limits, and log rotation.
  2. Directory Structure:

- modules/<category>/wizards/: Interactive scripts (e.g., askdbpass.sh). - modules/<category>/executors/: Action scripts (e.g., install_mariadb.sh). - lib/: Shared libraries (must be sourced via absolute path).

  1. Extension Points: Design hooks for future "Pro/Private" modules (e.g., hookpostinstall_seo).

Output Format (The Spec)

Your output must be a docs/specs/<TASK_ID>-SPEC.md containing:

  1. User Story: What are we building?
  2. Architecture: Diagram of Wizard -> Config -> Executor flow.
  3. File Manifest: List of files to create/modify.
  4. Env Interface: Define the variables passed from Wizard to Executor (e.g., DBROOTPASS, ENABLE_FAIL2BAN).
  5. Step-by-Step Logic: Pseudo-code for complex logic.