Skip to content
Agentmart
Submit a tool

using-superpowers

Consult directly matching task playbooks before substantial design, debugging, implementation, review, or completion work without forcing full workflow…

Skill by BB-84C

IndexedMIT★ 1
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/bb-84c-using-superpowers

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Consult directly matching task playbooks before substantial design, debugging, implementation, review, or completion work without forcing full workflow execution.

Use cases: design, productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/using-superpowers

What's inside

Skill name
using-superpowers
Description
Consult directly matching task playbooks before substantial design, debugging, implementation, review, or completion work without forcing full workflow execution.
License
MIT
Instructions
73 lines in SKILL.md · view on GitHub

Files (4)

  • SKILL.md
  • antigravity-tools.md · references/
  • codex-tools.md · references/
  • pi-tools.md · references/

SKILL.md (first lines)

<SUBAGENT-STOP>
If you were dispatched as a subagent to execute a specific task, ignore this skill.
</SUBAGENT-STOP>

# Consulting Task Playbooks

## Core rule

Skills are task-specific playbooks. Explicit user invocation of a skill is
binding.

Before substantial design, investigation, debugging, implementation, review, or
completion work, inspect the available skills. If a skill directly matches the
task type, load it before acting.

Loading a skill means consulting its guidance, not following every step verbatim.
Apply the parts relevant to the task; exact step-by-step execution is not
mandatory unless the skill explicitly marks a safety, permission, or verification
gate as binding.

## Select and consult

1. Read the request and available project instructions.
2. Scan the available skills for a direct task-type match.
3. Load the single most relevant workflow skill before substantial work begins.
4. Apply the guidance and binding gates relevant to the real task.
5. Load another skill only when the task enters a distinct phase.

Start with at most one primary workflow skill. Load another only when the task
enters a distinct phase or a separate risk boundary requires it.

Direct questions, narrow lookups, and bounded low-risk mechanical changes may
proceed without loading a workflow skill.

## Common routes

- Unresolved product behavior, architecture, interaction design, or another
  load-bearing choice: consult brainstorming after inspecting the project.
- Unclear root cause, repeated failure, or cross-component behavior: consult
  systematic-debugging before selecting a repair.
- A behavior change with a meaningful automated seam: consult
  test-driven-development before implementation.
- Multi-round implementation, context compaction, or worker handoff: consult
  writing-plans.
- Review feedback that needs evaluation before implementation: consult
  receiving-code-review.
- Delegating implementation from an approved plan: consult
  subagent-driven-development.
- A substantive completion claim: consult verification-before-completion.

These are task-type matches, not a fixed sequence.

## Delegated work

Workers executing a bounded delegated task use the supplied task contract. They
do not need to re-run this session-start selection step unless the task itself
requires a safety, verification, or other specialized workflow.

## Platform adaptation

If your harness appears here, read its reference file for tool translation only:

- Codex: `references/codex-tools.md`
- Pi: `references/pi-tools.md`
- Antigravity: `references/antigravity-tools.md`

## User instructions

Direct user instructions, repository instructions, and project policy take
precedence over this skill. Preserve user authority at approval, permission, and
repository-operation boundaries.

From BB-84C/superpowers-lite, MIT.

More from BB-84C/superpowers-lite

Signals

Stars
1
Forks
1
Open issues
0
Last commit (this folder)
Jul 15, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from BB-84C

Cite this page

using-superpowers skill. Agent Mart. https://agentmart.dev/skills/bb-84c-using-superpowers (last updated 2026-10-24)