kdy1-scripts

In the author's own words, from each skill's SKILL.md.

• add-issue

  Evidence-driven GitHub issue creation for confirmed bugs and decision-complete future work, with evidence-backed metadata where authorized. Start only on explicit human `$add-issue` invocation or an authorized candidate handoff from human-invoked MainQA or Static Analysis; never select this skill automatically from bug reports, feature ideas, TODOs, URLs, issue references, or task similarity. Cla…

  skills/add-issue/SKILL.md on GitHub (opens in a new tab)

• bulk-watch

  Watch a user-specified source and apply one common prompt to existing and newly discovered items through per-item worktree chats in the current project, automatically archiving chats that no longer need attention. Continue on a configurable heartbeat, defaulting to five minutes. Start on explicit human invocation or an authorized skill handoff.

  skills/bulk-watch/SKILL.md on GitHub (opens in a new tab)

• bulk

  Apply a common prompt to a finite text list of items through per-item worktree chats in the current project, coordinating results and automatically archiving chats that no longer need attention. Start on explicit human invocation or an authorized skill handoff.

  skills/bulk/SKILL.md on GitHub (opens in a new tab)

• create-human-reviewed-pr

  Publish human-reviewed changes as a non-draft GitHub pull request with AI-use disclosure, a human-reviewer mention, and available harness, model, and reasoning-effort details. Use only when the human explicitly invokes $create-human-reviewed-pr; never select it automatically from general PR requests, repository changes, or task similarity, and never start it through another skill, agent, or autom…

  skills/create-human-reviewed-pr/SKILL.md on GitHub (opens in a new tab)

• create-pr

  Publish intended changes as a GitHub pull request with verified Closes references for fully resolved issues and Refs for partial or related work, then attach the result. Use when an authorized task includes PR creation or another skill delegates its PR publishing step.

  skills/create-pr/SKILL.md on GitHub (opens in a new tab)

• list-good-prs

  List open pull requests in the current GitHub repository that are CLEAN or UNSTABLE and have a +1 reaction from chatgpt-codex-connector[bot]. Use when the user asks for good or Codex-approved PRs; this skill only reports them and never changes repository state.

  skills/list-good-prs/SKILL.md on GitHub (opens in a new tab)

• main-qa

  Run parallel web-app QA for functional and usability problems with GPT-5.6 Luna xhigh subagents and explicitly delegate discovered problems to $add-issue as they are found, recording verified candidates during QA. Start only when the human directly invokes $main-qa or explicitly requests MainQA. Test the supplied checklist exclusively, or discover and test the whole app when no checklist is suppl…

  skills/main-qa/SKILL.md on GitHub (opens in a new tab)

• manage-stacked-prs

  Modify existing GitHub stacked PRs and restack and verify affected upper layers with gh stack. Apply when editing a PR in an existing stack or repairing its alignment. Preserve unrelated lower PRs; fix a lower layer only when its confirmed defect blocks the requested work. Stack creation and merging remain separate workflows.

  skills/manage-stacked-prs/SKILL.md on GitHub (opens in a new tab)

• merge-pr

  Squash-merge GitHub pull requests immediately or enable auto-merge with the default commit subject when the user explicitly invokes this skill, requests a merge, or authorizes a handoff within that scope.

  skills/merge-pr/SKILL.md on GitHub (opens in a new tab)

• redesign-ui

  Select automatically only in Plan Mode for product UI changes, including layout, styling, components, content, interaction, navigation, or responsive behavior. Outside Plan Mode, do not select automatically; explicit `$redesign-ui` invocation requires a switch to Plan Mode before proposal work. Plan and visualize the redesign, then implement the approved design in the same conversation after leav…

  skills/redesign-ui/SKILL.md on GitHub (opens in a new tab)

• repair-pr

  One-shot GitHub pull request repair workflow for the current or specified repository. Use when asked to resolve merge conflicts, handle unresolved chatgpt-codex-connector[bot] feedback, fix failing CI, commit each distinct repair problem separately, push once, and resolve handled bot review threads. Post no PR comments by default, except optional English explanations of objectively incorrect bot…

  skills/repair-pr/SKILL.md on GitHub (opens in a new tab)

• review-full

  Run a sustained three-reviewer GitHub PR review and publish one consolidated COMMENT review after every reviewer exhausts new findings. Use when the user requests Review Full or this repeated-review-and-publish workflow, not for ordinary code reviews.

  skills/review-full/SKILL.md on GitHub (opens in a new tab)

• slop-fix-batch

  Fix a fixed batch of open repository issues selected by human-supplied conditions using parallel subagents, group related fixes, and publish stacked PRs with gh stack. Start only when the human explicitly invokes $slop-fix-batch.

  skills/slop-fix-batch/SKILL.md on GitHub (opens in a new tab)

• slop-fix-issue

  Generate AI slop to implement one GitHub issue, open and verify a non-draft PR with a Closes reference, and stop; close already-resolved issues with supporting evidence instead.

  skills/slop-fix-issue/SKILL.md on GitHub (opens in a new tab)

• slop-fix-repo-issues

  Collect open issues matching required user-specified conditions in the current repository once, then use $bulk to delegate each issue to $slop-fix-issue through PR creation. Start only when the human explicitly invokes $slop-fix-repo-issues.

  skills/slop-fix-repo-issues/SKILL.md on GitHub (opens in a new tab)

• slop-maintain-repo-prs

  Periodically scan my open PRs targeting the current repository's default branch, dispatch one-shot $repair-pr runs in GPT-6 Luna worktree chats with xhigh reasoning effort, recover incomplete repairs automatically, and archive successful repair chats. Start only when the human explicitly invokes $slop-maintain-repo-prs.

  skills/slop-maintain-repo-prs/SKILL.md on GitHub (opens in a new tab)

• static-analysis

  Analyze all code in a user-specified scope, or the whole repository when scope is omitted, and hand each confirmed bug to $add-issue immediately. Use for requested comprehensive static bug analysis; issue recording requires a direct human invocation of $static-analysis or an explicit request to run Static Analysis.

  skills/static-analysis/SKILL.md on GitHub (opens in a new tab)

• write-blog-post

  Collect the user's blog material, suggest topics to develop, and rephrase and reorder supplied content into a draft only on request. Use only when a human explicitly invokes $write-blog-post or names Write Blog Post; never select it automatically from a general blog-writing request.

  skills/write-blog-post/SKILL.md on GitHub (opens in a new tab)

• write-comment

  Use only when the user explicitly invokes Write Comment or $write-comment. Draft publication-ready public comments in Plan Mode for GitHub issues and PRs, code reviews, forums, and social posts. Include the exact comment text and destination in the final plan, then publish the approved text unchanged when execution is authorized outside Plan Mode. Default to English unless the user explicitly req…

  skills/write-comment/SKILL.md on GitHub (opens in a new tab)

• write-marketing-copy

  Rephrase a topic, notes, announcement, or draft into three platform-native marketing posts for LinkedIn, X, and Threads. Use when explicitly invoked to deliver English LinkedIn and X copy plus formal-polite Korean Threads copy.

  skills/write-marketing-copy/SKILL.md on GitHub (opens in a new tab)

• write-prd

  Product requirements decision capture and GitHub issue creation. Use only when a human explicitly invokes `$write-prd`; never select this skill automatically from feature, planning, PRD, or issue-creation task similarity. When explicitly invoked, define or finalize a PRD, turn a feature idea into a decision-complete GitHub issue, and resolve applicable product, technical, operational, rollout, ob…

  skills/write-prd/SKILL.md on GitHub (opens in a new tab)

• write-ste

  Write or revise human-readable technical text with practical ASD-STE100 principles. Use automatically for PR and issue titles and bodies, review comments, technical documentation, and code explanations, including technical chat replies. Preserve the text's language and meaning; apply clarity principles to languages other than English. Exclude ordinary conversation and marketing copy.

  skills/write-ste/SKILL.md on GitHub (opens in a new tab)

No skill matches that filter.

Similar skills

• superpowersby obra An agentic skills framework & software development methodology that works. 300k stars · 15 skills
• skillsby mattpocock Skills for Real Engineers. Straight from my .agents directory. 293k stars · 30 skills
• ECCby affaan-m The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Cod… 280k stars · 30 skills
• reaby morluto Reverse engineer anything with agents, from app behavior down to native binaries. 270k stars · 1 skill
• andrej-karpathy-skillsby multica-ai A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls. 220k stars · 1 skill
• skillsby anthropics Public repository for Agent Skills 181k stars · 19 skills