Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Superstack Teach

ASecurity

The human wants to understand — "walk me through this", "explain at my level", "teach me" — or a handover where the maintainer is the human. Skip for one-line answers and stranger-facing docs (a documentation task, not teaching).

2 stars
0 votes
0 copies
1 views
Added 9/19/2026
ai-agentsgodocumentation

Works with

cli

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add debabsah/superstack --skill superstack-teach --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Superstack Teach?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Superstack Teach
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/debabsah-superstack-teach/badge)](https://www.skillsdirectory.com/skills/debabsah-superstack-teach)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: superstack-teach
description: The human wants to understand — "walk me through this", "explain at my level", "teach me" — or a handover where the maintainer is the human. Skip for one-line answers and stranger-facing docs (a documentation task, not teaching).
---

# superstack-teach

A verified claim is not an understood system. Every other record in this method flows toward the model or the file; this one flows toward the human — the goal is *their* durable mental model, not another document.

The stranger-facing-docs deferral routes to a doc-generation tool only when one really exists here — look for it in the skill listing and say what you found; with none installed, superstack-ship's docs-as-done owns the document while this skill owns the human's understanding.

## The session

1. **Calibrate first.** Open `.superstack/toured.md` if it exists and start where the last tour ended — then one question about what they already know or use (not a quiz, a placement). Pitch the register there, and hold it: every term of art gets one plain-word gloss or stays out (jargon that outruns the owner is this method's most user-visible recorded failure).
2. **Walk the actual artifact, not an abstraction.** Open the real files; trace **one real flow end to end** — the request, the click, the query — before any architecture talk. When the claim is about *behavior* — what the code does when run — **enter the modality: run the flow and show its output** (superstack-experiential's rule); narrating from memory of what you wrote is the provenance rule failing against your yesterday-self. Name the load-bearing walls and *why* they're load-bearing; decision records (`docs/decisions/`, if the project keeps them) are the why-source — cite them rather than reconstructing rationale from memory.
3. **Check-backs, not lectures.** After each chunk, one question that only has an answer if the chunk landed — "what would break if we deleted X?" beats "does that make sense?" (which always gets yes). A wrong answer means re-teach *differently*, not louder — the register was wrong, not the listener.
4. **Name the honest boundary.** What this walkthrough deliberately skipped, and where it lives — a tour that implies completeness teaches a false map.

## The record

Append one line to `.superstack/toured.md` (step 1 is its reader):

```
- <date> <topic> — depth: <overview|working|deep>
```

So the next teaching session starts where this one ended instead of at zero. Expiry toward doubt applies: a tour ~90 days old is a *stale* map — offer a refresh, don't assume retention.

## Boundary

Docs-as-done (superstack-ship) writes for a stranger with no context; teach writes for *this* human's current model of the system. Producing a doc as a side effect is fine; producing *only* a doc means teaching didn't happen.

Attribution

debabsahdebabsah
View sourceMore from debabsah →
SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

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 (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1066601 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

693161 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

651 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →