Skip to content
Back to skills

Tails Help

ASecurity

Quick-reference card for all LLM_Tails modes, skills, and commands. One-shot display. Trigger: /tails-help, "tails help", "how do I use tails".

  • 3 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 9, 2026
ai-agentsgogitapi

Works with

  • api

Security analysis

A100/100

Scanned October 9, 2026

npx -y skills add savadks95/LLM_Tails --skill tails-help --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Tails Help?

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

Security grade badge for Tails Help
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/savadks95-tails-help/badge)](https://www.skillsdirectory.com/skills/savadks95-tails-help)

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: tails-help
description: >
  Quick-reference card for all LLM_Tails modes, skills, and commands.
  One-shot display. Trigger: /tails-help, "tails help", "how do I use tails".
---

# LLM_Tails Help

Display this reference card when invoked. One-shot, do NOT change mode.

## The Four Pillars

| # | Pillar | What it does |
|---|--------|-------------|
| 1 | **Understand first** | Read the codebase before editing. Trace callers, callees, conventions. Check the wiki. |
| 2 | **Minimal correct code** | The laziness ladder: YAGNI → stdlib → native → one-line → minimum. |
| 3 | **Never hallucinate** | Verify APIs, paths, configs before using them. Confidence markers: `[verified]`, `[likely]`, `[uncertain]`. |
| 4 | **Enforce consistency** | Match naming, patterns, error handling, imports, tests to what the project already does. |

## Levels

| Level | Trigger | What changes |
|-------|---------|-------------|
| **Lite** | `/tails lite` | Advisory: names the better alternative, user picks. |
| **Full** | `/tails` | All four pillars enforced. Default. |
| **Ultra** | `/tails ultra` | Maximum strictness. Every claim needs `[verified]` or explicit `[uncertain]`. |

## Skills

| Skill | Trigger | What it does |
|-------|---------|-------------|
| **tails** | `/tails` | The four pillars, always-on. |
| **tails-review** | `/tails-review` | Diff review through all four lenses. |
| **tails-audit** | `/tails-audit` | Whole-repo audit: bloat, hallucination risks, consistency, architecture. |
| **tails-wiki-init** | `/tails-wiki-init` | Scaffold a project wiki from the current codebase. |
| **tails-wiki-ingest** | `/tails-wiki-ingest` | Process a new source into the wiki. |
| **tails-wiki-lint** | `/tails-wiki-lint` | Health-check the wiki against current code. |
| **tails-debt** | `/tails-debt` | Harvest `tails:` shortcut comments into a tracked ledger. |
| **tails-help** | `/tails-help` | This card. |

## Comment markers

- `tails:` — deliberate simplification with ceiling and upgrade path.
- `tails: inconsistent pattern` — noted pattern conflict, following the prevalent one.

## Deactivate

Say "stop tails" or "normal mode". Resume with `/tails`.
`/tails off` also works.

## Compatibility

LLM_Tails is compatible with Ponytail (they can coexist — Ponytail governs
code minimalism, Tails adds architecture awareness, anti-hallucination, and
consistency on top).

## More

Full docs: https://github.com/savadks95/LLM_Tails

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.