diff --git a/.agents/skills/find-skills/SKILL.md b/.agents/skills/find-skills/SKILL.md deleted file mode 100644 index a41bdd0..0000000 --- a/.agents/skills/find-skills/SKILL.md +++ /dev/null @@ -1,141 +0,0 @@ ---- -name: find-skills -description: Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill. ---- - -# Find Skills - -This skill helps you discover and install skills from the open agent skills ecosystem. - -## When to Use This Skill - -Use this skill when the user: - -- Asks "how do I do X" where X might be a common task with an existing skill -- Says "find a skill for X" or "is there a skill for X" -- Asks "can you do X" where X is a specialized capability -- Expresses interest in extending agent capabilities -- Wants to search for tools, templates, or workflows -- Mentions they wish they had help with a specific domain (design, testing, deployment, etc.) - -## What is the Skills CLI? - -The Skills CLI (`npx skills`) is the package manager for the open agent skills ecosystem. Skills are modular packages that extend agent capabilities with specialized knowledge, workflows, and tools. - -**Key commands:** - -- `npx skills find [query] [--owner ]` - Search for skills interactively or by keyword, optionally scoped to a GitHub owner -- `npx skills add ` - Install a skill from GitHub or other sources -- `npx skills update` - Update all installed skills - -**Browse skills at:** https://skills.sh/ - -## How to Help Users Find Skills - -### Step 1: Understand What They Need - -When a user asks for help with something, identify: - -1. The domain (e.g., React, testing, design, deployment) -2. The specific task (e.g., writing tests, creating animations, reviewing PRs) -3. Whether this is a common enough task that a skill likely exists - -### Step 2: Check the Leaderboard First - -Before running a CLI search, check the [skills.sh leaderboard](https://skills.sh/) to see if a well-known skill already exists for the domain. The leaderboard ranks skills by total installs, surfacing the most popular and battle-tested options. - -For example, top skills for web development include: -- `vercel-labs/agent-skills` — React, Next.js, web design (100K+ installs each) -- `anthropics/skills` — Frontend design, document processing (100K+ installs) - -### Step 3: Search for Skills - -If the leaderboard doesn't cover the user's need, run the find command: - -```bash -npx skills find [query] [--owner ] -``` - -For example: - -- User asks "how do I make my React app faster?" → `npx skills find react performance` -- User asks "can you help me with PR reviews?" → `npx skills find pr review` -- User asks "I need to create a changelog" → `npx skills find changelog` - -### Step 4: Verify Quality Before Recommending - -**Do not recommend a skill based solely on search results.** Always verify: - -1. **Install count** — Prefer skills with 1K+ installs. Be cautious with anything under 100. -2. **Source reputation** — Official sources (`vercel-labs`, `anthropics`, `microsoft`) are more trustworthy than unknown authors. -3. **GitHub stars** — Check the source repository. A skill from a repo with <100 stars should be treated with skepticism. - -### Step 5: Present Options to the User - -When you find relevant skills, present them to the user with: - -1. The skill name and what it does -2. The install count and source -3. The install command they can run -4. A link to learn more at skills.sh - -Example response: - -``` -I found a skill that might help! The "react-best-practices" skill provides -React and Next.js performance optimization guidelines from Vercel Engineering. -(185K installs) - -To install it: -npx skills add vercel-labs/agent-skills@react-best-practices - -Learn more: https://skills.sh/vercel-labs/agent-skills/react-best-practices -``` - -### Step 6: Offer to Install - -If the user wants to proceed, you can install the skill for them: - -```bash -npx skills add -g -y -``` - -The `-g` flag installs globally (user-level) and `-y` skips confirmation prompts. - -## Common Skill Categories - -When searching, consider these common categories: - -| Category | Example Queries | -| --------------- | ---------------------------------------- | -| Web Development | react, nextjs, typescript, css, tailwind | -| Testing | testing, jest, playwright, e2e | -| DevOps | deploy, docker, kubernetes, ci-cd | -| Documentation | docs, readme, changelog, api-docs | -| Code Quality | review, lint, refactor, best-practices | -| Design | ui, ux, design-system, accessibility | -| Productivity | workflow, automation, git | - -## Tips for Effective Searches - -1. **Use specific keywords**: "react testing" is better than just "testing" -2. **Try alternative terms**: If "deploy" doesn't work, try "deployment" or "ci-cd" -3. **Check popular sources**: Many skills come from `vercel-labs/agent-skills` or `ComposioHQ/awesome-claude-skills` - -## When No Skills Are Found - -If no relevant skills exist: - -1. Acknowledge that no existing skill was found -2. Offer to help with the task directly using your general capabilities -3. Suggest the user could create their own skill with `npx skills init` - -Example: - -``` -I searched for skills related to "xyz" but didn't find any matches. -I can still help you with this task directly! Would you like me to proceed? - -If this is something you do often, you could create your own skill: -npx skills init my-xyz-skill -``` diff --git a/.claude-plugin/marketplace.json b/.claude-plugin/marketplace.json index befa77c..84cd09f 100644 --- a/.claude-plugin/marketplace.json +++ b/.claude-plugin/marketplace.json @@ -1,7 +1,7 @@ { "$schema": "https://anthropic.com/claude-code/marketplace.schema.json", "name": "killallgit", - "description": "Killallgit workflow skills for Claude Code and Codex.", + "description": "Killallgit workflow skills.", "owner": { "name": "archae0pteryx", "url": "https://github.com/killallgit" diff --git a/.claude/skills/find-skills b/.claude/skills/find-skills deleted file mode 120000 index 0b17e35..0000000 --- a/.claude/skills/find-skills +++ /dev/null @@ -1 +0,0 @@ -../../.agents/skills/find-skills \ No newline at end of file diff --git a/.cursor-plugin/marketplace.json b/.cursor-plugin/marketplace.json new file mode 100644 index 0000000..ae78c58 --- /dev/null +++ b/.cursor-plugin/marketplace.json @@ -0,0 +1,112 @@ +{ + "name": "killallgit", + "owner": { + "name": "archae0pteryx" + }, + "metadata": { + "description": "Killallgit workflow skills.", + "version": "5.5.0" + }, + "plugins": [ + { + "name": "architectural-review", + "source": "./skills/architectural-review", + "description": "Review a codebase's architecture. Finds hotspots in git history, walks them for friction, writes an HTML report of deepening candidates with before/after diagrams, proposes glossary terms, and suggests beads for the work. Words after the command name a direction (a module, a subsystem, a pain point).", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "catch-up", + "source": "./skills/catch-up", + "description": "Rebuild a human's lost context on the work in flight after reconciling the current checkout's dangling changes \u2014 commit what belongs to an open PR, discard what already landed, sync the branch \u2014 then summarize the state of play, decisions, live PRs and tickets, and the single next step. Use when the user says \"catch me up\", \"where were we\", \"what's the state of this\", \"get back on track\", or invokes `/catch-up` after time away.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "end-session", + "source": "./skills/end-session", + "description": "Bring this session's branch up to date, land or discard its work, and archive the session.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "git-janitor", + "source": "./skills/git-janitor", + "description": "Clean up local Git, only when the user runs /git-janitor. Inspect uncommitted and unpublished work in every worktree first, discard work that is superseded or over 60 days old when justified, then remove merged and stale local worktrees and branches. Record what was discarded and why.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "independent-review", + "source": "./skills/independent-review", + "description": "Get the current branch reviewed by a reviewer that did not write it. Opens a PR if none exists, waits for the repo's CI reviewers (CodeRabbit, PR-Agent) when it has any, otherwise runs a fresh-context local review agent, then resolves the findings. Use when the user asks for an independent review, or when another skill calls for one.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "lets-finish", + "source": "./skills/lets-finish", + "description": "Land finished work. Open the PR if there is none, wait for CI, get it independently reviewed, merge, update the tickets and beads it relates to, then clean up the branch and worktree. Use when the user says \"let's finish\", \"lets finish\", \"wrap up\", \"finish up\", \"ship it\", or invokes `/lets-finish`. Words after the command skip or stop steps, such as \"don't review\" or \"stop after CI\", \"and release\" or \"and deploy\" adds the project's release after the merge. Not for a question about status.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "restate", + "source": "./skills/restate", + "description": "Restate the request and scope before acting, then wait for approval", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "session-eject", + "source": "./skills/session-eject", + "description": "Save all work before usage runs out \u2014 commit, push, update or open draft PRs, and stop. `now` skips waiting for subagents.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "setup-agents-md", + "source": "./skills/setup-agents-md", + "description": "Set up and review root AGENTS.md and CLAUDE.md. Merge shared instructions from both into AGENTS.md without losing unique or tool-generated guidance, simplify the combined instructions, verify them against the repo, and make CLAUDE.md import AGENTS.md with @AGENTS.md. Create missing files as needed. Use for \"setup agents md\", \"check the agent docs\", \"check project\", \"dedupe CLAUDE.md and AGENTS.md\", or /setup-agents-md.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "wait-for-action", + "source": "./skills/wait-for-action", + "description": "Wait for CI, PR checks, a review, or a workflow run in one blocking call. Use before a merge (\"is it ready to merge\", wait for the PR to be green, CI, then the CodeRabbit or PR-Agent review), after a push, merge or workflow dispatch, or when a run failed and you need its log. Replaces any hand-written `gh run watch`, `gh pr checks --watch` or sleep loop.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "killall-skills", + "source": "./skills", + "description": "Helpful agent skills.", + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0", + "skills": [ + "./architectural-review", + "./catch-up", + "./end-session", + "./git-janitor", + "./independent-review", + "./lets-finish", + "./restate", + "./session-eject", + "./setup-agents-md", + "./wait-for-action" + ] + } + ] +} diff --git a/.github/plugin/marketplace.json b/.github/plugin/marketplace.json new file mode 100644 index 0000000..1753f2b --- /dev/null +++ b/.github/plugin/marketplace.json @@ -0,0 +1,153 @@ +{ + "name": "killallgit", + "owner": { + "name": "archae0pteryx" + }, + "metadata": { + "description": "Killallgit workflow skills.", + "version": "5.5.0" + }, + "plugins": [ + { + "name": "architectural-review", + "source": "./", + "description": "Review a codebase's architecture. Finds hotspots in git history, walks them for friction, writes an HTML report of deepening candidates with before/after diagrams, proposes glossary terms, and suggests beads for the work. Words after the command name a direction (a module, a subsystem, a pain point).", + "strict": false, + "skills": [ + "./skills/architectural-review" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "catch-up", + "source": "./", + "description": "Rebuild a human's lost context on the work in flight after reconciling the current checkout's dangling changes \u2014 commit what belongs to an open PR, discard what already landed, sync the branch \u2014 then summarize the state of play, decisions, live PRs and tickets, and the single next step. Use when the user says \"catch me up\", \"where were we\", \"what's the state of this\", \"get back on track\", or invokes `/catch-up` after time away.", + "strict": false, + "skills": [ + "./skills/catch-up" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "end-session", + "source": "./", + "description": "Bring this session's branch up to date, land or discard its work, and archive the session.", + "strict": false, + "skills": [ + "./skills/end-session" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "git-janitor", + "source": "./", + "description": "Clean up local Git, only when the user runs /git-janitor. Inspect uncommitted and unpublished work in every worktree first, discard work that is superseded or over 60 days old when justified, then remove merged and stale local worktrees and branches. Record what was discarded and why.", + "strict": false, + "skills": [ + "./skills/git-janitor" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "independent-review", + "source": "./", + "description": "Get the current branch reviewed by a reviewer that did not write it. Opens a PR if none exists, waits for the repo's CI reviewers (CodeRabbit, PR-Agent) when it has any, otherwise runs a fresh-context local review agent, then resolves the findings. Use when the user asks for an independent review, or when another skill calls for one.", + "strict": false, + "skills": [ + "./skills/independent-review" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "lets-finish", + "source": "./", + "description": "Land finished work. Open the PR if there is none, wait for CI, get it independently reviewed, merge, update the tickets and beads it relates to, then clean up the branch and worktree. Use when the user says \"let's finish\", \"lets finish\", \"wrap up\", \"finish up\", \"ship it\", or invokes `/lets-finish`. Words after the command skip or stop steps, such as \"don't review\" or \"stop after CI\", \"and release\" or \"and deploy\" adds the project's release after the merge. Not for a question about status.", + "strict": false, + "skills": [ + "./skills/lets-finish" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "restate", + "source": "./", + "description": "Restate the request and scope before acting, then wait for approval", + "strict": false, + "skills": [ + "./skills/restate" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "session-eject", + "source": "./", + "description": "Save all work before usage runs out \u2014 commit, push, update or open draft PRs, and stop. `now` skips waiting for subagents.", + "strict": false, + "skills": [ + "./skills/session-eject" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "setup-agents-md", + "source": "./", + "description": "Set up and review root AGENTS.md and CLAUDE.md. Merge shared instructions from both into AGENTS.md without losing unique or tool-generated guidance, simplify the combined instructions, verify them against the repo, and make CLAUDE.md import AGENTS.md with @AGENTS.md. Create missing files as needed. Use for \"setup agents md\", \"check the agent docs\", \"check project\", \"dedupe CLAUDE.md and AGENTS.md\", or /setup-agents-md.", + "strict": false, + "skills": [ + "./skills/setup-agents-md" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "wait-for-action", + "source": "./", + "description": "Wait for CI, PR checks, a review, or a workflow run in one blocking call. Use before a merge (\"is it ready to merge\", wait for the PR to be green, CI, then the CodeRabbit or PR-Agent review), after a push, merge or workflow dispatch, or when a run failed and you need its log. Replaces any hand-written `gh run watch`, `gh pr checks --watch` or sleep loop.", + "strict": false, + "skills": [ + "./skills/wait-for-action" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + }, + { + "name": "killall-skills", + "source": "./", + "description": "Helpful agent skills.", + "strict": false, + "skills": [ + "./skills/architectural-review", + "./skills/catch-up", + "./skills/end-session", + "./skills/git-janitor", + "./skills/independent-review", + "./skills/lets-finish", + "./skills/restate", + "./skills/session-eject", + "./skills/setup-agents-md", + "./skills/wait-for-action" + ], + "category": "workflow", + "homepage": "https://github.com/killallgit/killall-skills", + "version": "5.5.0" + } + ] +} diff --git a/.gitignore b/.gitignore index 9ffc0db..4bf36c0 100644 --- a/.gitignore +++ b/.gitignore @@ -19,3 +19,5 @@ _old # Experiment scratch output tmp/ +.venv +___* diff --git a/README.md b/README.md index f2dca5a..972e63d 100644 --- a/README.md +++ b/README.md @@ -18,6 +18,23 @@ codex plugin marketplace add killallgit/killall-skills codex plugin add killall-skills@killallgit ``` +GitHub Copilot CLI: + +``` +copilot plugin marketplace add killallgit/killall-skills +copilot plugin install killall-skills@killallgit +``` + +Cursor, from the agent chat: + +``` +/plugin marketplace add killallgit/killall-skills +``` + +Then install `killall-skills` from the Marketplace tab. The Cursor CLI can add +the marketplace with `cursor-agent plugin marketplace add`, and the install +step stays in that tab. + `killall-skills` installs every skill. To install one, use its name instead, for example `git-janitor@killallgit`. Pick either the bundle or individual skills, not both. @@ -32,8 +49,12 @@ Claude Code runs a skill as `/name`; Codex runs it as `$name`. ## Development -`.claude-plugin/marketplace.json` is generated from each skill's `SKILL.md` -frontmatter and `marketplace.config.json`; Claude Code and Codex both read it. +`task manifests` regenerates the catalogs from each skill's `SKILL.md` +frontmatter and `scripts/marketplace.config.json`: + +- `.claude-plugin/marketplace.json` for Claude Code, Codex, and `npx skills` +- `.cursor-plugin/marketplace.json` for Cursor +- `.github/plugin/marketplace.json` for GitHub Copilot CLI After adding, renaming, or editing a skill description: ``` @@ -47,7 +68,11 @@ them into that skill's entry and into the `killall-skills` bundle. The bundle is listed last so the skills CLI groups all skills under `Killall Skills` instead of repeating each skill name as a group heading. -To release, run one of these from anywhere in the checkout. The release runs in +Before releasing, update your checkout to the latest `main`: the release runner +is loaded from your checkout, not the temporary worktree. Older runners do not +support the relocated `scripts/marketplace.config.json`. + +Then run one of these from anywhere in the checkout. The release runs in a temporary worktree off `origin/main`, so your checkout is never switched. It runs the tests and `generate_manifests.py --check` first, and on a failed push it removes the tag it created: @@ -58,7 +83,7 @@ task tag:minor task tag:major ``` -It bumps `version` in `marketplace.config.json`, regenerates the manifest, +It bumps `version` in `scripts/marketplace.config.json`, regenerates the manifest, commits, tags `vX.Y.Z`, and pushes the commit and tag together. The release is built from `origin/main`, so uncommitted changes in your checkout are ignored and never released. diff --git a/scripts/generate_manifests.py b/scripts/generate_manifests.py index ac6e4af..3fdf0c6 100644 --- a/scripts/generate_manifests.py +++ b/scripts/generate_manifests.py @@ -74,7 +74,7 @@ def build_entry( return {**entry, **{key: value for key, value in optional.items() if value}} -def build_marketplace(config: dict, skills: tuple[Skill, ...]) -> dict: +def plugin_entries(config: dict, skills: tuple[Skill, ...]) -> list[dict]: bundle = build_entry( config, config["bundle"]["name"], @@ -88,11 +88,61 @@ def build_marketplace(config: dict, skills: tuple[Skill, ...]) -> dict: ) for skill in skills ] + return [*singles, bundle] + + +def build_marketplace(config: dict, skills: tuple[Skill, ...]) -> dict: return { "$schema": "https://anthropic.com/claude-code/marketplace.schema.json", "name": config["name"], "description": config["description"], "owner": config["owner"], + "plugins": plugin_entries(config, skills), + } + + +def host_metadata(config: dict) -> dict: + metadata = {"description": config["description"]} + if version := config.get("version"): + metadata["version"] = version + return metadata + + +def host_owner(config: dict) -> dict: + return {"name": config["owner"]["name"]} + + +def build_copilot_marketplace(config: dict, skills: tuple[Skill, ...]) -> dict: + return { + "name": config["name"], + "owner": host_owner(config), + "metadata": host_metadata(config), + "plugins": plugin_entries(config, skills), + } + + +def build_cursor_marketplace(config: dict, skills: tuple[Skill, ...]) -> dict: + def entry(name: str, description: str, source: str) -> dict: + plugin = { + "name": name, + "source": source, + "description": description, + "category": config["category"], + "homepage": config["homepage"], + } + if version := config.get("version"): + plugin["version"] = version + return plugin + + singles = [ + entry(skill.name, skill.description, f"./skills/{skill.name}") for skill in skills + ] + bundle = entry(config["bundle"]["name"], config["bundle"]["description"], "./skills") + bundle["skills"] = [f"./{skill.name}" for skill in skills] + return { + "name": config["name"], + "owner": host_owner(config), + "metadata": host_metadata(config), "plugins": [*singles, bundle], } @@ -102,13 +152,22 @@ def render(data: dict) -> str: def generated_files(root: Path) -> dict[Path, str]: - config = json.loads((root / "marketplace.config.json").read_text()) + config = json.loads((root / "scripts" / "marketplace.config.json").read_text()) skills = read_skills(root / "skills") - return { + cursor = build_cursor_marketplace(config, skills) + files = { root / ".claude-plugin" / "marketplace.json": render( build_marketplace(config, skills) ), + root / ".cursor-plugin" / "marketplace.json": render(cursor), + root / ".github" / "plugin" / "marketplace.json": render( + build_copilot_marketplace(config, skills) + ), } + for plugin in cursor["plugins"]: + manifest = {key: plugin[key] for key in ("name", "description", "skills") if key in plugin} + files[root / plugin["source"] / ".cursor-plugin" / "plugin.json"] = render(manifest) + return files def drifted_files(files: dict[Path, str]) -> list[Path]: @@ -126,13 +185,14 @@ def write_files(files: dict[Path, str]) -> None: def main(argv: list[str]) -> int: - files = generated_files(Path(__file__).resolve().parent.parent) + root = Path(__file__).resolve().parent.parent + files = generated_files(root) if "--check" not in argv: write_files(files) return 0 drifted = drifted_files(files) for path in drifted: - print(f"out of date: {path.name}", file=sys.stderr) + print(f"out of date: {path.relative_to(root)}", file=sys.stderr) return 1 if drifted else 0 diff --git a/marketplace.config.json b/scripts/marketplace.config.json similarity index 81% rename from marketplace.config.json rename to scripts/marketplace.config.json index 6453ce8..86528ad 100644 --- a/marketplace.config.json +++ b/scripts/marketplace.config.json @@ -1,6 +1,6 @@ { "name": "killallgit", - "description": "Killallgit workflow skills for Claude Code and Codex.", + "description": "Killallgit workflow skills.", "version": "5.5.0", "owner": { "name": "archae0pteryx", diff --git a/scripts/release.py b/scripts/release.py index e6e5b36..7f91ab7 100644 --- a/scripts/release.py +++ b/scripts/release.py @@ -11,7 +11,12 @@ SEMVER = re.compile(r"\A(\d+)\.(\d+)\.(\d+)\Z") LEVELS = ("patch", "minor", "major") -RELEASE_FILES = [".claude-plugin/marketplace.json", "marketplace.config.json"] +RELEASE_FILES = [ + ".claude-plugin/marketplace.json", + ".cursor-plugin/marketplace.json", + ".github/plugin/marketplace.json", + "scripts/marketplace.config.json", +] def bump_version(current: str, level: str) -> str: @@ -70,7 +75,7 @@ def output(step: list[str], cwd: Path) -> str: def release_from(work: Path, level: str) -> str: for step in check_steps(): run(step, work) - config_path = work / "marketplace.config.json" + config_path = work / "scripts" / "marketplace.config.json" config = json.loads(config_path.read_text()) config["version"] = bump_version(config["version"], level) tag = f"v{config['version']}" diff --git a/skills-lock.json b/skills-lock.json deleted file mode 100644 index 535dc0a..0000000 --- a/skills-lock.json +++ /dev/null @@ -1,11 +0,0 @@ -{ - "version": 1, - "skills": { - "find-skills": { - "source": "vercel-labs/skills", - "sourceType": "github", - "skillPath": "skills/find-skills/SKILL.md", - "computedHash": "913b9d37d0d54047dd65222bb8c67b2bf04e3cb87dcad1729068d7a8b2c8c396" - } - } -} diff --git a/skills/.cursor-plugin/plugin.json b/skills/.cursor-plugin/plugin.json new file mode 100644 index 0000000..7c78041 --- /dev/null +++ b/skills/.cursor-plugin/plugin.json @@ -0,0 +1,16 @@ +{ + "name": "killall-skills", + "description": "Helpful agent skills.", + "skills": [ + "./architectural-review", + "./catch-up", + "./end-session", + "./git-janitor", + "./independent-review", + "./lets-finish", + "./restate", + "./session-eject", + "./setup-agents-md", + "./wait-for-action" + ] +} diff --git a/skills/architectural-review/.cursor-plugin/plugin.json b/skills/architectural-review/.cursor-plugin/plugin.json new file mode 100644 index 0000000..2ba006d --- /dev/null +++ b/skills/architectural-review/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "architectural-review", + "description": "Review a codebase's architecture. Finds hotspots in git history, walks them for friction, writes an HTML report of deepening candidates with before/after diagrams, proposes glossary terms, and suggests beads for the work. Words after the command name a direction (a module, a subsystem, a pain point)." +} diff --git a/skills/catch-up/.cursor-plugin/plugin.json b/skills/catch-up/.cursor-plugin/plugin.json new file mode 100644 index 0000000..043dc50 --- /dev/null +++ b/skills/catch-up/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "catch-up", + "description": "Rebuild a human's lost context on the work in flight after reconciling the current checkout's dangling changes \u2014 commit what belongs to an open PR, discard what already landed, sync the branch \u2014 then summarize the state of play, decisions, live PRs and tickets, and the single next step. Use when the user says \"catch me up\", \"where were we\", \"what's the state of this\", \"get back on track\", or invokes `/catch-up` after time away." +} diff --git a/skills/end-session/.cursor-plugin/plugin.json b/skills/end-session/.cursor-plugin/plugin.json new file mode 100644 index 0000000..fe39910 --- /dev/null +++ b/skills/end-session/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "end-session", + "description": "Bring this session's branch up to date, land or discard its work, and archive the session." +} diff --git a/skills/git-janitor/.cursor-plugin/plugin.json b/skills/git-janitor/.cursor-plugin/plugin.json new file mode 100644 index 0000000..8b07348 --- /dev/null +++ b/skills/git-janitor/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "git-janitor", + "description": "Clean up local Git, only when the user runs /git-janitor. Inspect uncommitted and unpublished work in every worktree first, discard work that is superseded or over 60 days old when justified, then remove merged and stale local worktrees and branches. Record what was discarded and why." +} diff --git a/skills/independent-review/.cursor-plugin/plugin.json b/skills/independent-review/.cursor-plugin/plugin.json new file mode 100644 index 0000000..c22ea1f --- /dev/null +++ b/skills/independent-review/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "independent-review", + "description": "Get the current branch reviewed by a reviewer that did not write it. Opens a PR if none exists, waits for the repo's CI reviewers (CodeRabbit, PR-Agent) when it has any, otherwise runs a fresh-context local review agent, then resolves the findings. Use when the user asks for an independent review, or when another skill calls for one." +} diff --git a/skills/lets-finish/.cursor-plugin/plugin.json b/skills/lets-finish/.cursor-plugin/plugin.json new file mode 100644 index 0000000..affb085 --- /dev/null +++ b/skills/lets-finish/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "lets-finish", + "description": "Land finished work. Open the PR if there is none, wait for CI, get it independently reviewed, merge, update the tickets and beads it relates to, then clean up the branch and worktree. Use when the user says \"let's finish\", \"lets finish\", \"wrap up\", \"finish up\", \"ship it\", or invokes `/lets-finish`. Words after the command skip or stop steps, such as \"don't review\" or \"stop after CI\", \"and release\" or \"and deploy\" adds the project's release after the merge. Not for a question about status." +} diff --git a/skills/restate/.cursor-plugin/plugin.json b/skills/restate/.cursor-plugin/plugin.json new file mode 100644 index 0000000..30dc751 --- /dev/null +++ b/skills/restate/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "restate", + "description": "Restate the request and scope before acting, then wait for approval" +} diff --git a/skills/session-eject/.cursor-plugin/plugin.json b/skills/session-eject/.cursor-plugin/plugin.json new file mode 100644 index 0000000..08bda0a --- /dev/null +++ b/skills/session-eject/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "session-eject", + "description": "Save all work before usage runs out \u2014 commit, push, update or open draft PRs, and stop. `now` skips waiting for subagents." +} diff --git a/skills/setup-agents-md/.cursor-plugin/plugin.json b/skills/setup-agents-md/.cursor-plugin/plugin.json new file mode 100644 index 0000000..17396b5 --- /dev/null +++ b/skills/setup-agents-md/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "setup-agents-md", + "description": "Set up and review root AGENTS.md and CLAUDE.md. Merge shared instructions from both into AGENTS.md without losing unique or tool-generated guidance, simplify the combined instructions, verify them against the repo, and make CLAUDE.md import AGENTS.md with @AGENTS.md. Create missing files as needed. Use for \"setup agents md\", \"check the agent docs\", \"check project\", \"dedupe CLAUDE.md and AGENTS.md\", or /setup-agents-md." +} diff --git a/skills/wait-for-action/.cursor-plugin/plugin.json b/skills/wait-for-action/.cursor-plugin/plugin.json new file mode 100644 index 0000000..d75137f --- /dev/null +++ b/skills/wait-for-action/.cursor-plugin/plugin.json @@ -0,0 +1,4 @@ +{ + "name": "wait-for-action", + "description": "Wait for CI, PR checks, a review, or a workflow run in one blocking call. Use before a merge (\"is it ready to merge\", wait for the PR to be green, CI, then the CodeRabbit or PR-Agent review), after a push, merge or workflow dispatch, or when a run failed and you need its log. Replaces any hand-written `gh run watch`, `gh pr checks --watch` or sleep loop." +} diff --git a/tests/test_generate_manifests.py b/tests/test_generate_manifests.py index 40cf3e4..d69650c 100644 --- a/tests/test_generate_manifests.py +++ b/tests/test_generate_manifests.py @@ -5,6 +5,8 @@ from generate_manifests import ( Skill, + build_copilot_marketplace, + build_cursor_marketplace, build_marketplace, drifted_files, generated_files, @@ -32,7 +34,9 @@ def write_skill(root: Path, dirname: str, frontmatter: str) -> None: def write_repo(root: Path) -> None: write_skill(root, "alpha", "name: alpha\ndescription: first") - (root / "marketplace.config.json").write_text(json.dumps(CONFIG)) + scripts = root / "scripts" + scripts.mkdir() + (scripts / "marketplace.config.json").write_text(json.dumps(CONFIG)) def test_parse_skill_reads_plain_description(): @@ -118,14 +122,84 @@ def test_skill_entry_description_comes_from_frontmatter(): assert alpha["description"] == "first" -def test_generated_files_cover_only_the_marketplace(tmp_path): +def test_generated_files_cover_claude_cursor_and_copilot(tmp_path): write_repo(tmp_path) assert set(generated_files(tmp_path)) == { - tmp_path / ".claude-plugin" / "marketplace.json" + tmp_path / ".claude-plugin" / "marketplace.json", + tmp_path / ".cursor-plugin" / "marketplace.json", + tmp_path / ".github" / "plugin" / "marketplace.json", + tmp_path / "skills" / "alpha" / ".cursor-plugin" / "plugin.json", + tmp_path / "skills" / ".cursor-plugin" / "plugin.json", } +def catalog(tmp_path, relative): + return json.loads(generated_files(tmp_path)[tmp_path / relative]) + + +def test_cursor_catalog_points_each_plugin_at_its_skill_directory(tmp_path): + write_repo(tmp_path) + write_skill(tmp_path, "beta", "name: beta\ndescription: second") + plugins = catalog(tmp_path, ".cursor-plugin/marketplace.json")["plugins"] + + assert [(p["name"], p["source"]) for p in plugins] == [ + ("alpha", "./skills/alpha"), + ("beta", "./skills/beta"), + ("bundle", "./skills"), + ] + assert all("strict" not in p for p in plugins) + assert all("skills" not in p for p in plugins[:-1]) + assert plugins[-1]["skills"] == ["./alpha", "./beta"] + + +def test_cursor_sources_have_manifests_and_discover_every_skill(tmp_path): + write_repo(tmp_path) + write_skill(tmp_path, "beta", "name: beta\ndescription: second") + write_files(generated_files(tmp_path)) + marketplace = json.loads((tmp_path / ".cursor-plugin" / "marketplace.json").read_text()) + + for plugin in marketplace["plugins"]: + source = tmp_path / plugin["source"] + manifest = json.loads((source / ".cursor-plugin" / "plugin.json").read_text()) + assert manifest["name"] == plugin["name"] + assert manifest["description"] == plugin["description"] + if plugin["name"] == "bundle": + assert manifest["skills"] == ["./alpha", "./beta"] + assert all((source / path / "SKILL.md").is_file() for path in manifest["skills"]) + else: + assert "skills" not in manifest + assert not (source / "skills").exists() + assert (source / "SKILL.md").is_file() + + +def test_cursor_manifests_inherit_version_from_the_marketplace(tmp_path): + write_repo(tmp_path) + config_path = tmp_path / "scripts" / "marketplace.config.json" + config_path.write_text(json.dumps({**CONFIG, "version": "1.2.3"})) + files = generated_files(tmp_path) + marketplace = json.loads(files[tmp_path / ".cursor-plugin" / "marketplace.json"]) + + for plugin in marketplace["plugins"]: + manifest = json.loads(files[tmp_path / plugin["source"] / ".cursor-plugin" / "plugin.json"]) + assert plugin["version"] == "1.2.3" + assert "version" not in manifest + + +def test_copilot_catalog_selects_skills_from_the_repo_root(tmp_path): + write_repo(tmp_path) + write_skill(tmp_path, "beta", "name: beta\ndescription: second") + marketplace = catalog(tmp_path, ".github/plugin/marketplace.json") + + assert marketplace["metadata"]["description"] == CONFIG["description"] + assert [p["skills"] for p in marketplace["plugins"]] == [ + ["./skills/alpha"], + ["./skills/beta"], + ["./skills/alpha", "./skills/beta"], + ] + assert {(p["source"], p["strict"]) for p in marketplace["plugins"]} == {("./", False)} + + def test_drifted_files_is_empty_after_write(tmp_path): write_repo(tmp_path) files = generated_files(tmp_path) @@ -182,6 +256,16 @@ def test_bundle_merges_the_hooks_of_every_skill_by_event(): } +def test_copilot_entries_keep_hooks_and_cursor_entries_leave_them_in_the_skill(): + hooked = (Skill("alpha", "first", hooks=ALPHA_HOOKS), SKILLS[1]) + + copilot = build_copilot_marketplace(CONFIG, hooked)["plugins"] + cursor = build_cursor_marketplace(CONFIG, hooked)["plugins"] + + assert copilot[0]["hooks"] == ALPHA_HOOKS + assert all("hooks" not in plugin for plugin in cursor) + + def test_entries_without_hooks_have_no_hooks_key(): plugins = build_marketplace(CONFIG, SKILLS)["plugins"] diff --git a/tests/test_release.py b/tests/test_release.py index 34f72ec..1b566c1 100644 --- a/tests/test_release.py +++ b/tests/test_release.py @@ -78,8 +78,8 @@ def checkout(tmp_path, monkeypatch): "category": "c", "bundle": {"name": "b", "description": "d"}, } - (clone / "marketplace.config.json").write_text(json.dumps(config)) (clone / "scripts").mkdir() + (clone / "scripts" / "marketplace.config.json").write_text(json.dumps(config)) (clone / "scripts" / "generate_manifests.py").write_text(generator.read_text()) skill = clone / "skills" / "one" skill.mkdir(parents=True) diff --git a/tests/test_render_report.py b/tests/test_render_report.py index d1b7715..c120f7c 100644 --- a/tests/test_render_report.py +++ b/tests/test_render_report.py @@ -1,5 +1,4 @@ import pytest - from render_report import Bead, InvalidFindings, bead_plan, render FINDINGS = { @@ -21,7 +20,9 @@ "bead": {"title": "Deepen Order intake", "type": "task", "priority": 2}, } ], - "glossary": [{"term": "Order", "definition": "A request to buy.", "avoid": ["Purchase"]}], + "glossary": [ + {"term": "Order", "definition": "A request to buy.", "avoid": ["Purchase"]} + ], } @@ -33,7 +34,10 @@ def test_render_escapes_candidate_text(): def test_render_rejects_an_unknown_strength(): - bad = {**FINDINGS, "candidates": [{**FINDINGS["candidates"][0], "strength": "maybe"}]} + bad = { + **FINDINGS, + "candidates": [{**FINDINGS["candidates"][0], "strength": "maybe"}], + } with pytest.raises(InvalidFindings): render(bad, date="2026-11-08") @@ -56,14 +60,39 @@ def candidate(**changes): {"repo": "acme"}, {**FINDINGS, "candidates": [candidate(), candidate()]}, {**FINDINGS, "candidates": [candidate(), candidate(id='2">