agentleFS
Sign inStart free
salmandevsalmandev/everything-gemini/skills/superpowers/SKILL.md

superpowers

Skill2 starsChanged 28 days agoApache-2.0View on GitHub

Enforces disciplined agentic software development methodology. Use when embarking on non-trivial coding tasks to conduct interactive specification gathering, create step-by-step red-green TDD implementation plans, orchestrate scoped subagents, and ensure zero-drift verification before committing changes.

15 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use context7, to-prd and 2 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • JetBrainsSkill

    context7

    Up-to-date library and framework documentation via Context7 MCP. Use when setup, API, or version-specific questions require current documentation.

    Also in 4 projects that use this

  • GoogleSkill

    to-prd

    Turn the current conversation into a PRD and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

    Also in 3 projects that use this

  • GoogleSkill

    tdd

    Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

    Also in 3 projects that use this

  • +1 more from google

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 4 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 12 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When a user requests a new feature, refactor, or complex fix:

    line 68
  2. Distinctive:

    Never write code immediately: Stop and review the existing architecture.

    line 69
  3. Very specific:

    Ask clarifying questions via interactive prompts (ask_question or direct user inquiry) regarding ambiguities, performance budgets, and …

    line 71
  4. Distinctive:

    Never skip Red: Verify the test fails before writing the implementation to prevent false-positive tests.

    line 110
  5. Very specific:

    Never commit unverified code: Run the full validation suite (npm test or bash scripts/validate.sh).

    line 111
  6. Very specific:

    Safety Shield Verification: Ensure all shell operations adhere to hooks/safety-shield and

    line 112
  7. Very specific:

    Atomic Git Commit: Commit the completed step with a semantic message (e.g.

    line 113
  8. Distinctive:

    Use the native CLI command to manage the Superpowers workflow:

    line 119

“Common” is measured: kept by at least 6 in 10 of the 12 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Agentic Superpowers Software Methodology (superpowers)

The Superpowers skill enforces a disciplined, high-fidelity software engineering methodology across autonomous coding agents. It prevents unstructured hallucinations, monolithic untested diffs, and prompt drift by turning the agent into a meticulous Staff Software Engineer.

Inspired by the workflow pioneered by Jesse Vincent in obra/superpowers and customized for Google Antigravity & Gemini 3.8, this skill organizes execution into a 5-phase deterministic cycle.

1. Architectural Workflow: The 5-Phase Superpowers Cycle

┌─────────────────────────────────────────────────────────────────────────────┐
│                     THE AGENTIC SUPERPOWERS METHODOLOGY                     │
│                                                                             │
│   ┌─────────────────────────────────────────────────────────────────────┐   │
│   │ Phase 1: Spec Gathering & Requirements Interview                     │   │
│   │    • Do NOT touch production code yet.                              │   │
│   │    • Ask clarifying questions; tease out edge cases & constraints.  │   │
│   │    • Draft concise specification in `.superpowers/spec.md`.         │   │
│   └──────────────────────────────────┬──────────────────────────────────┘   │
│                                      │ (User Sign-off)                      │
│                                      ▼                                      │
│   ┌─────────────────────────────────────────────────────────────────────┐   │
│   │ Phase 2: Step-by-Step TDD Implementation Plan                       │   │
│   │    • Deconstruct into bite-sized atomic steps (15-30 min each).     │   │
│   │    • Strict Red/Green TDD: specify failing tests before code.       │   │
│   │    • Enforce YAGNI (You Aren't Gonna Need It) & DRY principles.     │   │
│   │    • Store actionable checklist in `.superpowers/plan.md`.          │   │
│   └──────────────────────────────────┬──────────────────────────────────┘   │
│                                      │                                      │
│                                      ▼                                      │
│   ┌─────────────────────────────────────────────────────────────────────┐   │
│   │ Phase 3: Subagent Orchestration & Role Specialization               │   │
│   │    • Dispatch everything-gemini's 18 scoped subagents:              │   │
│   │      - planner: breaks down dependency graph                        │   │
│   │      - cve-mitigator & security-auditor: supply-chain & secrets     │   │
│   │      - resilience-engineer: chaos boundaries & timeout failover     │   │
│   │      - code-reviewer: reviews incremental diffs before commit       │   │
│   └──────────────────────────────────┬──────────────────────────────────┘   │
│                                      │                                      │
│                                      ▼                                      │
│   ┌─────────────────────────────────────────────────────────────────────┐   │
│   │ Phase 4: Atomic Execution & Red-Green-Refactor Loop                 │   │
│   │    • Red: Write failing unit/integration test.                      │   │
│   │    • Green: Write minimum necessary code to pass the test.          │   │
│   │    • Refactor: Clean up without breaking tests or adding unneeded features. │
│   └──────────────────────────────────┬──────────────────────────────────┘   │
│                                      │                                      │
│                                      ▼                                      │
│   ┌─────────────────────────────────────────────────────────────────────┐   │
│   │ Phase 5: Zero-Drift Closure & Lifecycle Hook Verification            │   │
│   │    • Run test suites (npm test / scripts/validate.sh).              │   │
│   │    • Gate with hooks (safety-shield, zero-trust-gate, budget-guard).│   │
│   │    • Produce atomic git commit and detailed verification artifact.  │   │
│   └─────────────────────────────────────────────────────────────────────┘   │
└─────────────────────────────────────────────────────────────────────────────┘

2. Phase 1: Clarifying Specification & Requirements Interview

When a user requests a new feature, refactor, or complex fix:

  1. Never write code immediately. Stop and review the existing architecture.
  2. Conduct the Clarification Interview:
    • Ask clarifying questions via interactive prompts (ask_question or direct user inquiry) regarding ambiguities, performance budgets, and backward compatibility.
  3. Generate .superpowers/spec.md:
    • Problem statement and user goals.
    • Explicit non-goals (what we will NOT build).
    • Core API contracts and data schemas.
    • Success metrics (e.g., latency, test coverage, zero regression).

3. Phase 2: Implementation Planning & Red/Green TDD

Once the spec is confirmed:

  1. Break down the task into 10 to 20 numbered atomic steps.
  2. For each step, define:
    • Test First (Red): The exact test file and test assertion that must fail initially.
    • Implementation (Green): The exact file to edit and the minimal implementation.
    • Verification: The command to run to verify that the test now passes.
  3. Apply the YAGNI rule: Remove speculative hooks, premature abstractions, and unused utility functions.

4. Phase 3: Subagent Delegation Matrix

Leverage everything-gemini's 18 scoped subagents for specialist review:

Phase / ConcernDelegated SubagentResponsibilities
Task Planningagents/planner.mdMaps dependency graph and orders atomic execution steps.
Dependency & CVE Auditagents/cve-mitigator.mdAudits libraries, lockfiles, and virtual environments for zero-day CVEs.
Security & Secretsagents/security-auditor.mdAudits code for hardcoded secrets, injection vectors, and privilege leaks.
Chaos & Error Boundariesagents/resilience-engineer.mdInjects synthetic timeouts, 429 shocks, and verifies circuit breakers.
Diff Reviewagents/code-reviewer.mdReviews incremental git diffs against style guides and DRY principles.
Documentation & Releaseagents/release-engineer.mdUpdates CHANGELOG, semver tagging, and release notes.

5. Phase 4 & 5: TDD Execution & Zero-Drift Closure

During implementation:

  • Never skip Red: Verify the test fails before writing the implementation to prevent false-positive tests.
  • Never commit unverified code: Run the full validation suite (npm test or bash scripts/validate.sh).
  • Safety Shield Verification: Ensure all shell operations adhere to hooks/safety-shield and hooks/zero-trust-gate.
  • Atomic Git Commit: Commit the completed step with a semantic message (e.g. feat(superpowers): add tdd verification harness).

6. CLI Command & Automation

Use the native CLI command to manage the Superpowers workflow:

# Initialize a Superpowers workflow directory in active workspace
everything-gemini superpowers init

# Check adherence to spec, TDD status, and test coverage
everything-gemini superpowers check

# Run the complete automated TDD validation loop
everything-gemini superpowers run --dry-run

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free