https://github.com/bitcoin21ideas/skills
Cross-agent skills for AI coding agents (Claude Code, Codex, Cursor, Copilot, Gemini), installable with gh skill install.
https://github.com/bitcoin21ideas/skills
agent-skills ai-agents anthropic claude claude-code claude-code-skills claude-skills codex cursor developer-tools gemini-cli github-copilot llm productivity
Last synced: about 14 hours ago
JSON representation
Cross-agent skills for AI coding agents (Claude Code, Codex, Cursor, Copilot, Gemini), installable with gh skill install.
- Host: GitHub
- URL: https://github.com/bitcoin21ideas/skills
- Owner: bitcoin21ideas
- License: mit
- Created: 2026-05-11T06:51:48.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2026-06-19T11:37:50.000Z (6 days ago)
- Last Synced: 2026-06-19T13:24:10.600Z (6 days ago)
- Topics: agent-skills, ai-agents, anthropic, claude, claude-code, claude-code-skills, claude-skills, codex, cursor, developer-tools, gemini-cli, github-copilot, llm, productivity
- Size: 2.63 MB
- Stars: 1
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Agents: AGENTS.md
Awesome Lists containing this project
README
# skills
[](LICENSE)
[](https://github.com/bitcoin21ideas/skills/releases/latest)
[](https://zapmeacoffee.com/npub10awzknjg5r5lajnr53438ndcyjylgqsrnrtq5grs495v42qc6awsj45ys7)
A collection of agent skills for Claude Code, Codex, Copilot, and other AI coding assistants.
> [!NOTE]
> These skills are designed to be useful across agents. Unknown frontmatter fields (like `disable-model-invocation`) are ignored per the agentskills.io standard, so Claude-Code-specific fields are safe to include in any compliant agent.
## Skills
| Skill | Description |
| --- | --- |
| [pressure-test](skills/pressure-test/README.md) | Interviews you relentlessly about a plan, branch by branch, and ends with a consolidated decision artifact you can optionally save to a `.decisions.md` file for to-plan. |
| [to-plan](skills/to-plan/README.md) | Synthesizes a decisions record and a repo read into a pure implementation plan on disk — file paths, modules, phase-gated slices, and anchored verification — shaped to be hardened by plan-clash. |
| [plan-clash](skills/plan-clash/README.md) | Hardens an existing implementation plan by looping a Codex (GPT) critic and a Claude (opus) adjudicator over it for N rounds — cross-model, fresh-eyes review that improves the plan against the real code. Runs on CLI subscriptions; improves a plan, never certifies one. |
| [commit](skills/commit/README.md) | Project-agnostic git commit workflow — triages secrets, runs the project's test/lint gate, groups changes, and writes conventional commits after you approve. Tailorable per project. |
## Install
> [!IMPORTANT]
> A skill is instructions your agent runs on your behalf — anyone can publish one, and GitHub doesn't verify them. Inspect before you install: `gh skill preview bitcoin21ideas/skills `, or read the skill's `SKILL.md` on GitHub. The skills here are MIT-licensed and meant to be auditable; apply the same caution to skills from anywhere.
Three ways, easiest first: the **GitHub CLI** (one command, cross-agent), **ask your agent**, or **by hand**.
### GitHub CLI — `gh skill` (recommended)
[`gh skill`](https://github.blog/changelog/2026-04-16-manage-agent-skills-with-github-cli/) (GitHub CLI ≥ 2.90.0) installs a skill into whichever agent you use — Claude Code, Copilot, Cursor, Codex, or Gemini CLI — resolving the right location for you and recording where it came from:
```sh
gh skill preview bitcoin21ideas/skills pressure-test # inspect first
gh skill install bitcoin21ideas/skills pressure-test # then install
```
Swap `pressure-test` for any skill in the table above.
### Ask your agent
No CLI? Hand your agent the skill's folder URL and ask it to install:
> I like this skill — install it in my project:
> https://github.com/bitcoin21ideas/skills/tree/main/skills/pressure-test
This needs an agent that can fetch a URL — **Claude Code** and **Gemini CLI** do by default; **Cursor** and **Copilot** may need web access turned on; **Codex CLI** needs a fetch tool (MCP) wired up. Tell it two things: a skill is a multi-file *folder* (not one file), and read the **raw** files (`raw.githubusercontent.com`), not the rendered `blob` page. On a model with no skills runtime (a bare Kimi / OpenAI-compatible endpoint), just paste `SKILL.md` into your system prompt.
### By hand
```sh
git clone https://github.com/bitcoin21ideas/skills.git
cd skills
```
Drop a skill folder where your agent looks. `.agents/skills/` is the vendor-neutral location honored by Codex, Cursor, Gemini, and Copilot:
```sh
ln -s "$PWD/skills/pressure-test" .agents/skills/pressure-test # project-level
ln -s "$PWD/skills/pressure-test" ~/.agents/skills/pressure-test # user-level
```
Prefer an agent-specific home?
| Agent | Project path | User path | Invoke |
| --- | --- | --- | --- |
| Claude Code | `.claude/skills/` | `~/.claude/skills/` | `/pressure-test` |
| Codex CLI | `.agents/skills/` | `~/.agents/skills/` | `$pressure-test` |
| Cursor | `.cursor/skills/` | `~/.cursor/skills/` | `/` menu |
| Copilot (VS Code) | `.github/skills/` | `~/.copilot/skills/` | `/pressure-test` |
| Gemini CLI | `.gemini/skills/` | `~/.gemini/skills/` | `/skills` |
Most agents also auto-load a skill when its description matches, so explicit invocation is optional.
## Cross-agent compatibility note
As of June 2026 the [Agent Skills standard](https://agentskills.io) is natively supported across the major coding agents (Claude Code, Codex, Cursor, Copilot, Gemini CLI, and dozens more), so these skills are portable without per-agent rewrites.
The `disable-model-invocation: true` frontmatter field prevents Claude Code and Copilot from auto-invoking a skill based on its description — the user must call it explicitly. Agents that don't recognize the field ignore it safely, falling back to description-based triggering.
Invocation mode is chosen per skill:
- **`pressure-test`** is explicit-only (`disable-model-invocation: true`) — you rarely want to be auto-grilled.
- **`plan-clash`** is explicit-only (`disable-model-invocation: true`) — each run spawns long, subscription-billed CLI subprocesses, so it should fire only when you ask for it.
- **`commit`** is both user- and model-invocable (no such field) — natural-language requests like "commit this" can trigger it, and it always proposes a plan and waits for approval before committing, so the model can never make a surprise commit.
Either way, skill descriptions are phrased with explicit user-action triggers ("when the user says...") so even agents that ignore the field don't fire them spuriously.
## Contributing
This repo is agent-first. The easiest way to add or fix a skill is to ask your coding agent: point it at [`AGENTS.md`](AGENTS.md), describe the change, and have it open a PR. See [`AGENTS.md`](AGENTS.md) for the layout and conventions. Issues and PRs welcome — from an agent or by hand.
## License
MIT — see [LICENSE](LICENSE).
> _Commits in this repo are made with its own [`commit`](skills/commit/README.md) skill._