Skip to content
Back to skills

Sanctuary Spec Kitty

ASecurity

Project Sanctuary-specific skill for Spec-Driven Development. Knows the project's constitution, safety rules, AUGMENTED.md best practices, and how the spec-kitty-plugin should be configured for this project.

  • 5 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added June 6, 2026
toolsrustbashgit

Works with

  • cli

Security analysis

A100/100

Scanned June 6, 2026

npx -y skills add richfrem/Project_Sanctuary --skill sanctuary-spec-kitty --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Sanctuary Spec Kitty?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Sanctuary Spec Kitty
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/richfrem-sanctuary-spec-kitty/badge)](https://www.skillsdirectory.com/skills/richfrem-sanctuary-spec-kitty)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: sanctuary-spec-kitty
description: "Project Sanctuary-specific skill for Spec-Driven Development. Knows the project's constitution, safety rules, AUGMENTED.md best practices, and how the spec-kitty-plugin should be configured for this project."
---

# Sanctuary Spec-Kitty Configuration

**Status:** Active
**Domain:** Project Sanctuary
**Depends on:** `spec-kitty-plugin` (generic spec-driven development framework)

## Purpose

This skill is the **Sanctuary-specific glue layer** that knows how to configure and use the generic `spec-kitty-plugin` for Project Sanctuary's development workflow.

The spec-kitty-plugin has two layers:
1. **Auto-synced commands** (from upstream CLI) — generic, overwritten on update
2. **AUGMENTED.md files** (our custom best practices) — project-specific, never overwritten

This skill documents what's Sanctuary-specific about our spec-kitty setup.

## Sanctuary-Specific Configuration

### Constitution
The project constitution lives at `.kittify/memory/constitution.md` and is synced to `.agent/rules/constitution.md`. It defines:
- The Human Gate (no state-changing ops without explicit approval)
- The Hybrid Workflow (Universal start → Learning Loop or Custom Flow)
- Zero Trust git policy (never commit to main, never push without approval)
- Tool Discovery enforcement (use query_cache.py, no grep for tools)

### AUGMENTED.md Files
These contain project-specific best practices that survive spec-kitty CLI updates:

| File | Sanctuary-Specific Content |
|---|---|
| `commands/spec-kitty-merge/AUGMENTED.md` | Pre-merge remote backup protocol, branch protection awareness (GH006 handling), kitty-specs conflict resolution |
| `commands/spec-kitty-implement/AUGMENTED.md` | Worktree discipline, absolute paths only, commit hygiene |
| `commands/spec-kitty-review/AUGMENTED.md` | Batch review protocol, dependency verification |
| `commands/spec-kitty-specify/AUGMENTED.md` | Doc co-authoring integration for specification quality |
| `commands/spec-kitty-plan/AUGMENTED.md` | Doc co-authoring integration for plan quality |

### Workflow SKILL.md
`skills/spec-kitty-workflow/SKILL.md` is the end-to-end guide with:
- Anti-simulation rules (proof = pasted command output)
- Deterministic closure protocol (review → accept → retrospective → merge → verify)
- Pre-merge remote backup (mandatory)
- Dual-Loop mode integration (Protocol 133)

### Orchestration Diagram
The end-to-end orchestration lifecycle—including `verify_workflow_state.py` gates and the downstream `intelligence-sync`—is mapped in our master diagram:
`resources/diagrams/standard-spec-kitty-workflow.mmd`

### Sync Plugin
To update when spec-kitty CLI upgrades:
```bash
# Invoke the sync-plugin skill
/spec-kitty.sync-plugin
```
This runs `sync_configuration.py`, then the agent reviews AUGMENTED.md files for staleness.

## Protected Files

The following files MUST NOT be deleted or corrupted during any sync or merge:
- `.agent/rules/constitution.md`
- `.agent/rules/standard-workflow-rules.md`
- `.agent/rules/01_PROCESS/*`
- `.agent/rules/02_OPERATIONS/*`
- `.agent/rules/03_TECHNICAL/*`

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.