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 Spike

ASecurity

A feasibility question, not a delivery — "can X even do Y", "is this viable", "how expensive would Z be" — where throwaway code is the cheapest probe. Skip when a read-only probe answers, when reasoning settles it (superstack-decide), and for work meant to be kept (superstack-scope).

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

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

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

Installs into .claude/skills of the current project.

Are you the author of Superstack Spike?

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

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

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

Download with Pro
Files
SKILL.md
---
name: superstack-spike
description: A feasibility question, not a delivery — "can X even do Y", "is this viable", "how expensive would Z be" — where throwaway code is the cheapest probe. Skip when a read-only probe answers, when reasoning settles it (superstack-decide), and for work meant to be kept (superstack-scope).
---

# superstack-spike

A spike buys an **answer**, not code. The code has no legal status; the answer does — a spike without its harvest was just deleted work.

## 1. The contract (one block, before building)

```
question: <falsifiable — what observation settles it>
timebox: <duration>
touches: <scratch only | real data | live service | money>
stop when: answered · timebox spent · <signal it's a dead end>
```

## 2. Build the cheapest thing that answers

Ship discipline is suspended **by declaration, and only when `touches: scratch only`**: no tests, no docs, no review; hardcode freely; T1 on a branch or scratch dir. A probe that reads real data, hits a live service, or spends money keeps its tier and its gates (the tier table in the superstack gateway skill) — **tier is set by blast radius, never by intent; declaring "spike" lowers nothing** (the method's rule: effort never lowers a tier's minimum). The suspension is legal because the contract above exists and step 3 is mandatory.

**Inside an ACTIVE campaign, a spike is off-frontier:** no receipt, no frontier movement; its harvest enters the plan as a write-ahead discovery line, and the campaign's red-before-green discipline resumes untouched.

## 3. The harvest (mandatory)

- **The answer, as a ledger line:** `Verified: <answer> — ran <probe> -> saw <result>` — or `PROVISIONAL: <best reading>` when the timebox ran out unclear (an honest "still unknown, here's what it cost to learn that" is a legal spike result). **Append it to `.superstack/claims-log`** like any shipped claim — an unrecorded `Verified:` can never be falsified.
- **The code's disposition, stated:** `discard` (default) or `rebuild properly`. Spike code never graduates by renaming — keeping it means re-entering through superstack-scope, where it earns tests and review like anything else.
- **What it cost** vs the timebox — appended to the claims-log line itself (`… — cost: <spent>/<boxed>`), so how long spikes here actually take is a readable base rate that tunes the next contract.

## 4. Route the answer somewhere durable

Retires a scope unknown → the task file (none open → the overlay). Changes a premise → the premise ledger (inception's `P<n>` block; entered standalone with no ledger → the overlay, as a dated fact). Spawns real work → superstack-queue (its grammar). A durable project fact → the overlay (`.superstack/project.md`). An answer that lives only in the chat dies with the session.

## Red flag

"The spike works — let me just clean it up and keep it." That's a build wearing a spike's exemption. Stop; re-enter through scope.

Attribution

debabsahdebabsah
View sourceMore from debabsah →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

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

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

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', ...

686011 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 →