Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
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
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Wizard

ASecurity

Use when asked to generate a bash wizard for provisioning, credentials, dashboards, migrations, or cutovers a human performs. Not for remote, credential, publish, deploy, or irreversible changes.

54 stars
0 votes
0 copies
0 views
Added 9/20/2026
ai-agentsshellbashdockergit

Works with

cli

Security Analysis

A100/100

Pro scans all 4 files and shows the line behind each finding

Scanned 9/20/2026

$npx -y skills add OutlineDriven/outline-driven-development --skill wizard --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Wizard?

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

Security grade badge for Wizard
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/outlinedriven-wizard-outline-driven-development/badge)](https://www.skillsdirectory.com/skills/outlinedriven-wizard-outline-driven-development)

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

Download with Pro
Files
SKILL.md
---
name: wizard
description: 'Use when asked to generate a bash wizard for provisioning, credentials, dashboards, migrations, or cutovers a human performs. Not for remote, credential, publish, deploy, or irreversible changes.'
---

# Wizard

## Contract

| Field | Bound contract |
|---|---|
| Trigger | User needs provisioning, credentials, dashboard, or migration steps that only a human can perform. |
| Authority | Reversible local: write only the wizard script to the named target path; delete the file to roll back. |
| Side effect | Writes the wizard script to the target path and makes it executable; verification is static only: the script is never executed. |
| Done | The wizard script exists at the target path, is executable, passes static verification, and every collected value is traced to its destination. |

## Inputs

1. **Procedure description** (required): what the wizard must accomplish: provisioning, credential setup, dashboard walkthrough, migration, or cutover.
2. **Target path** (optional): where to write the script; default `scripts/setup-<topic>.sh`.
3. **Repository context** (read automatically): `.env`, `.env.example`, `.env.*`, `README`, `docker-compose*`, framework configuration, `.github/workflows/*`.

## Procedure

### 1. Scope the procedure

Read the repository before asking anything: `.env`, `.env.example`, and `.env.*`; `README`; `docker-compose*`; framework configuration; `.github/workflows/*`. Treat every `secrets.*` and `vars.*` reference in a workflow as a value the wizard must produce. For a migration, read the current state, the target state, and every irreversible action between them.

Show the ordered stage list and the values each stage produces. Confirm it with the user. Done when: every stage has a name and order, every captured value has a known source, every captured value has a destination (`.env`, a GitHub secret, both, or nowhere for a pure-action stage), and every captured value is classified as secret or public.

### 2. Map the human journey

For each stage, write instructions a stranger can follow. Name the URL, the clicks, where the value appears, and which variable it fills. If the current UI or exact command is unknown, say so. Check the official docs or ask the user. Never invent a step. Done when: every stage has stranger-followable instructions naming the URL, clicks, value location, and target variable.

### 3. Author the script

Fail if a file already exists at the target path; do not overwrite. Copy `scripts/wizard-template.sh` to the target path. Replace the example with one `stage` per step in dependency order. Set `TOTAL_STAGES` and `TOTAL_MINUTES` to honest values because they drive the time-remaining display. Use the library helpers by contract per `references/library-helpers.md`. Done when: script is written with one stage per step, honest totals, and all helpers used correctly.

### 4. Verify and hand off

1. Run `bash -n <script>`.
2. Run `shellcheck <script>` when ShellCheck is available.
3. Run `chmod +x <script>`.
4. Static journey trace: parse the script without executing it. For every read cycle confirm the read variable has a named destination, every destination is a writable path or a named env-var export, and the script terminates with a clear success message and exit 0. If any read cycle is untraced, fail.
5. Trace every scoped value. Confirm it is captured and reaches its declared destination.
6. Confirm every `set_secret` name matches a `secrets.*` workflow reference exactly.
7. Tell the user how to run the script.

Done when: script passes `bash -n` and shellcheck (when available), is executable, every read cycle is traced to a destination with exit 0, every scoped value is traced to its destination, and every `set_secret` name matches its workflow reference.

Commit the wizard only when the user wants a repeatable setup path in the repository. Otherwise, treat it as ephemeral and delete it after the job is done.

## Failure and recovery
| Failure class | Behavior |
|---|---|
| Scope not confirmed | Stop before authoring. Re-present the stage list and wait. |
| Path collision | Fail before write. Report the existing path. Do not overwrite. |
| Value trace fails | Stop. Report which value has no capture or no destination. Do not hand off. |
| Untraced read cycle | Roll back the wizard file. Report the untraced block and the missing destination. |
| `bash -n` or shellcheck fails | Stop. Report the syntax error. Fix before handoff. |
| `set_secret` name mismatch | Stop. Report the secret name and the workflow reference. Reconcile before handoff. |
| `gh` unavailable | Record in `SKIPPED`. The script warns at runtime; the user sets the secret manually. |

Partial-result rule: a partially authored script is never handed off. Rollback: delete the script file if it has not been committed.

## Output
A self-contained bash wizard script at the target path containing the shared library, stages in dependency order, honest `TOTAL_STAGES` and `TOTAL_MINUTES`, and every scoped value captured and routed to its declared destination.

Attribution

OutlineDrivenOutlineDriven
View sourceSee grades on GitHubMore from OutlineDriven →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

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

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 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', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 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.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, 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.

741 votes
View all in ai-agents →