{
  "markdown": "# dmi_superpowers\n\nAn open-source, continuously distilled collection of the best skills and concepts for giving coding agents the discipline to write high-quality code — consolidating ideas from the superpowers plugin, Matt Pocock's skills, Ponytail, and Cursor's Thermos plugin into a single coherent methodology engine.\n\nContributions are welcome. See [AGENTS.md](AGENTS.md) for what belongs here and how to write a pull request.\n\n## What It Is\n\n`dmi_superpowers` packages 26 skills that guide coding agents through a disciplined workflow:\n\n**brainstorm → PRD → grill-with-docs → TSP → grill → TDD → review → finish**\n\nThe agent never just starts coding. It steps back, refines requirements, validates against domain language, plans carefully, writes tests first, reviews its own work, and closes out cleanly.\n\n## The 26 Skills\n\n### Kept from superpowers (core workflow)\n\n| Skill | Purpose |\n|---|---|\n| `brainstorming` | Socratic design refinement before any code |\n| `writing-plans` | Detailed implementation plans with task breakdown |\n| `executing-plans` | Batch execution with human checkpoints |\n| `dispatching-parallel-agents` | Concurrent subagent workflows |\n| `requesting-code-review` | Pre-review checklist and dispatch |\n| `receiving-code-review` | Responding to feedback with rigor |\n| `resolving-merge-conflicts` | Resolve an in-progress git merge/rebase conflict |\n| `using-git-worktrees` | Parallel development branches |\n| `finishing-a-development-branch` | Merge/PR decision workflow |\n| `creating-a-pull-request` | PR title/body for non-developer + developer audiences |\n| `subagent-driven-development` | Fast iteration with two-stage review |\n| `using-superpowers` | Introduction to the skills system |\n| `writing-skills` | Create new skills following best practices |\n\n### TDD-swapped (superpowers TDD replaced with tdd skill)\n\n| Skill | Purpose |\n|---|---|\n| `test-driven-development` | RED-GREEN-REFACTOR cycle (from tdd skill) |\n| `systematic-debugging` | 4-phase root cause process |\n| `verification-before-completion` | Ensure it's actually fixed |\n\n### Added from Matt Pocock's skills\n\n| Skill | Purpose |\n|---|---|\n| `codebase-design` | Shared vocabulary for designing deep modules — interfaces, seams, depth |\n| `domain-modeling` | Build and sharpen the domain model; record architectural decisions (ADRs) |\n| `grill-with-docs` | Challenge plans against CONTEXT.md and ADRs |\n| `grilling` | Relentlessly interview to stress-test a plan or design before building |\n| `improve-codebase-architecture` | Find deepening/refactor opportunities |\n| `say` | Communicate code findings in plain + technical pairs, and rewrite prose for human readers |\n| `to-prd` | Convert rough ideas to structured PRDs |\n\n### Merged / unified (debugging consolidated)\n\n| Skill | Purpose |\n|---|---|\n| `prototype` | Rapid prototyping skill |\n\n### Utility / mode skills\n\n| Skill | Purpose |\n|---|---|\n| `writing-code-comments` | Keep developer notes short — one plain sentence about intent or reason, not a walkthrough of the code |\n| `ponytail` | persistent minimalism mode: defaults every solution to the simplest thing that works (YAGNI, stdlib/native first, shortest diff), at intensity lite/full/ultra. Off by default; type `ponytail [level]` or `be lazy` to activate with plain text. |\n\n## Installation\n\nInstall dmi-superpowers separately for each harness you use.\n\n### Claude Code\n\n```bash\n/plugin install dmi-superpowers@dmi-marketplace\n```\n\nOr from this repository directly:\n\n```bash\n/plugin marketplace add HundredBillion/dmi_superpowers\n/plugin install dmi-superpowers@dmi-marketplace\n```\n\n### Antigravity\n\n```bash\nagy plugin install https://github.com/HundredBillion/dmi_superpowers\n```\n\n### Codex CLI\n\n- Open the plugin search interface:\n\n  ```bash\n  /plugins\n  ```\n\n- Search for dmi-superpowers and select `Install Plugin`.\n\n### Gemini CLI\n\n```bash\ngemini extensions install https://github.com/HundredBillion/dmi_superpowers\n```\n\nUpdate later:\n\n```bash\ngemini extensions update dmi-superpowers\n```\n\n### Kimi Code\n\n```text\n/plugins install https://github.com/HundredBillion/dmi_superpowers\n```\n\nDetailed docs: [docs/README.kimi.md](docs/README.kimi.md)\n\n### OpenCode\n\n```\nFetch and follow instructions from https://raw.githubusercontent.com/HundredBillion/dmi_superpowers/refs/heads/main/.opencode/INSTALL.md\n```\n\nDetailed docs: [docs/README.opencode.md](docs/README.opencode.md)\n\n### Pi\n\n```bash\npi install git:github.com/HundredBillion/dmi_superpowers\n```\n\nFor local development:\n\n```bash\npi -e /path/to/dmi_superpowers\n```\n\n## The Workflow\n\n1. **brainstorming** — Activates before writing code. Refines rough ideas through questions, explores alternatives, presents design in sections for validation. Saves design document.\n\n2. **to-prd** — Converts conversation output into a structured PRD. Produces `docs/PRDs/`.\n\n3. **grill-with-docs** — Challenges the PRD against CONTEXT.md and ADRs. Sharpens terminology, updates domain docs.\n\n4. **writing-plans** — Breaks work into bite-sized tasks (2–5 min each). Every task has exact file paths, interfaces, and verification steps. Produces `docs/TSPs/`.\n\n5. **grill** *(second pass)* — A second grill run validates the TSP against the domain model and prior decisions.\n\n6. **test-driven-development** — Enforces RED-GREEN-REFACTOR during implementation. Write failing test, watch it fail, write minimal code, pass, commit.\n\n7. **subagent-driven-development** / **executing-plans** — Dispatches fresh subagents per task with two-stage review (spec compliance then quality), or executes inline with human checkpoints.\n\n8. **requesting-code-review** — Reviews against plan, reports issues by severity. Critical issues block progress.\n\n9. **finishing-a-development-branch** — Verifies tests, presents options (merge/PR/keep/discard), cleans up.\n\n**Design documents live in `docs/PRDs/` and `docs/TSPs/`.**\n\n## Philosophy\n\n- **Test-Driven Development** — write tests first, always\n- **Systematic over ad-hoc** — process over guessing\n- **Complexity reduction** — simplicity as primary goal\n- **Evidence over claims** — verify before declaring success\n- **Domain language first** — grill against CONTEXT.md before building\n\n## License\n\nMIT — see LICENSE file for details.\n\n## Acknowledgements\n\n`dmi_superpowers` builds on ideas and workflows from several open-source projects:\n\n- [obra/superpowers](https://github.com/obra/superpowers) provides the core brainstorming, planning, TDD, debugging, review, and branch-completion workflow.\n- [mattpocock/skills](https://github.com/mattpocock/skills) provides the domain-modeling, architecture, PRD, grilling, communication, and prototyping skills consolidated here.\n- [DietrichGebert/ponytail](https://github.com/DietrichGebert/ponytail), created by Dietrich Gebert, is the source of the `ponytail` minimalism skill and persistence hook, adapted here under the MIT License.\n- Cursor's [Thermos plugin](https://github.com/cursor/plugins/tree/main/thermos), particularly `thermo-nuclear-code-quality-review`, inspired the structural-quality lens in `requesting-code-review`: code-judo reframing, large-file decomposition, abstraction quality, and aggressive simplification without changing behavior. The rubric originated in Cursor's built-in code-quality review prompt with additional constraints requested by Sualeh Asif; Eric Zakariasson contributed the public plugin version. Cursor's plugin repository is available under the MIT License.\n\nThese are influences incorporated into `dmi-superpowers`; users install this plugin as one package rather than installing each source plugin separately.\n",
  "bytes": 7599,
  "sha": "6db321d185b879f754f4bb9f9d97cf3e66e7868823113cd38f006d66a516d698",
  "repo_slug": "hundredbillion/dmi_superpowers",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_hundredbillion_dmi_superpowers_1f52e786/readme"
}