All categories
Development
Programming, frameworks, implementation, frontend, backend, and app development
- 70,054
- 2,919
Security grades appear on each card once the skill has been scanned. Newly imported skills may briefly show without a grade until the backfill job runs.
Open in full browserBrowse development skills
Showing 18,697–18,720 of 70,054 skills
- FinishRun the final cumulative adversarial review now — regardless of final_review, even with phases outstanding — and complete the activation if it passes.Votes: 0GitHub stars: 2
- ConfigRead or write the adversarial-review-loop configuration — every key's resolved value and which layer set it, or set/unset one key in the user config or (with --repo) the repository config.Votes: 0GitHub stars: 2
- AcceptManually approve the current working tree for this phase, overriding a review loop that will not converge, without leaving the mode.Votes: 0GitHub stars: 2
- Adversarial ReviewAdversarial code review of changes to adversarial-review-loop: working tree, staged diff, branch, commit range, or PR. Hunts for review-gate bypasses, fail-open hooks, wrong Git snapshots, unsafe state transitions, shell injection, reviewer-contract parsing bugs and tests that are not isolated, then reports ranked findings. Use when the user asks to review changes, a diff, a PR, a branch or a commit, check work before committing, assess merge readiness, or poke holes in an implementation.Votes: 0GitHub stars: 2
- Claude CliDelegate a scoped review or file-editing task from Codex to the official local Claude Code CLI when the user asks to use Claude, get Claude's second opinion, or hand Claude implementation work. Not for ordinary tasks or autonomous repeated review loops.Votes: 0GitHub stars: 2
- Serve GameLaunch the local dev server for Pixel Princess Platformer (the Kaplay browser platformer in this repo) so it can be played in a browser or driven with Playwright. Use when asked to run, start, serve, or open the game, or before browser-testing it.Votes: 0GitHub stars: 3
- Regen AssetsRegenerate all sprites, tiles, backgrounds and audio for Pixel Princess Platformer from the pixel-art pipeline. Use when asked to regenerate, rebuild, or refresh the game's generated assets.Votes: 0GitHub stars: 3
- Mobile CheckRun the iPhone-landscape emulation checks for Pixel Princess Platformer (audio unlock, touch controls, landscape fit, install hint) and review the screenshots. Use when asked to test the mobile or iOS experience.Votes: 0GitHub stars: 3
- DeployDeploy Pixel Princess Platformer to Vercel production (or a preview) and verify the live site. Use when asked to deploy, ship, publish, or push the game live.Votes: 0GitHub stars: 3
- TestsTesting doctrine, test layout conventions, and where each kind of test lives in the anywh repoVotes: 0GitHub stars: 4
- YtcliRead and change Yandex Tracker issues from the command line — get, search, count, comment, create, update, transition, worklogs, checklists and links, plus queues, boards and sprints, fields, templates, projects, portfolios, goals and attachments — and the organisation's Yandex Wiki pages, comments, files, grids and access. Use whenever a task mentions Tracker, Яндекс Трекер, an issue key like PROJ-123, a queue, a Tracker URL, Yandex Wiki, Яндекс Вики, or a wiki.yandex.ru address, and when a ...Votes: 0GitHub stars: 2
- Pctk Workflow Create WorkdirCreates a task folder under `.claude/__workdir/<task>/` (from the template embedded in this file's own \"Template do index.md\" section) together with a git worktree per repository the task touches, under `_worktrees/<task>/<repo>/`, and links the current conversation into the folder's `index.md` as `` - <título> (`<session_id>`) `` — the title read automatically from the session's own transcript, never typed by hand. Also covers two narrower cases: adding one more repository's worktree to a ...Votes: 0GitHub stars: 9
- Pctk Authoring Create SkillCreates a new domain skill under `domain/<nome>/SKILL.mdx`, shaped like a function: frontmatter (signature) + `<Context>` (required, purpose) + `<InputFormat>` (optional, parameters) + body (required, the logic) + `<OutputFormat>` (optional, return shape). Triggers on an explicit human instruction to create a new skill in `domain/` (e.g. \"cria uma skill em domain/ que faça X\", \"quero uma skill nova pra Y\") — never on ambient mention of skills. Out of scope: creating a `<Rule>` or `<Block>...Votes: 0GitHub stars: 9
- Pctk Authoring Create RuleCreates a new domain rule or convention under `domain/<nome>.mdx`, wrapped in `<Rule>`: `paths` (optional, frontmatter — scopes it to matching files; omitted, it loads every session as a convention) + body (required, the rule text itself). Triggers on an explicit human instruction to create a new rule or convention in `domain/` (e.g. \"cria uma rule em domain/ que exija X\", \"quero uma convention pra Y\") — never on ambient mention of rules. Out of scope: creating a `<Skill>` or `<Block>` do...Votes: 0GitHub stars: 9
- Pctk Authoring Create BlockCreates a new domain block under `domain/<nome>.mdx`, wrapped in `<Block>`: no props, just body (required, the shared fragment's content). Triggers on an explicit human instruction to create a new block in `domain/` (e.g. \"cria um block em domain/ com X\", \"quero um trecho reutilizável pra Y\") — never on ambient mention of blocks. Out of scope: creating a `<Skill>` or `<Rule>` domain artifact, and editing a block that already exists in `domain/`.Votes: 0GitHub stars: 9
- Test Driven DevelopmentDrives development with tests. Use when implementing any logic, fixing any bug, or changing any behavior. Use when you need to prove that code works, when a bug report arrives, or when you're about to modify existing functionality.Votes: 0GitHub stars: 5
- Spec Driven DevelopmentCreates specs before coding. Use when starting a new project, feature, or significant change and no specification exists yet. Use when requirements are unclear, ambiguous, or only exist as a vague idea.Votes: 0GitHub stars: 5
- Planning And Task BreakdownBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.Votes: 0GitHub stars: 5
- Incremental ImplementationDelivers changes incrementally. Use when implementing any feature or change that touches more than one file. Use when you're about to write a large amount of code at once, or when a task feels too big to land in one step.Votes: 0GitHub stars: 5
- GrillingGrill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.Votes: 0GitHub stars: 5
- Definition Of DoneProject-level completion standard checklist. Every task must meet these criteria before completion—includes code quality, test coverage, documentation, security, and ship-readiness checks.Votes: 0GitHub stars: 5
- Debugging And Error RecoveryGuides systematic root-cause debugging. Use when tests fail, builds break, behavior doesn't match expectations, or you encounter any unexpected error. Use when you need a systematic approach to finding and fixing the root cause rather than guessing.Votes: 0GitHub stars: 5
- Codoop Ux WalkthroughSimulate a real task from a chosen user persona and write an evidence-backed, non-blocking experience report. Use after building a user-facing feature, when reviewing a runnable product flow, or when product owners want user-perspective improvement ideas without changing code automatically.Votes: 0GitHub stars: 5
- Codoop TicketDesign work tickets (PRD → Spec → Plan) in three stages. It first researches comparable products, recommends a project-specific direction, and grills requirements one decision at a time before orchestrating PM and Architect agents. Its terminal responsibility is promoting an approved ticket to pending/; it never executes the ticket.Votes: 0GitHub stars: 5