- Skills
- 26
- 仓库
- 1
- 职业覆盖
- 0
- 场景覆盖
- 0
- Stars
- ★ 0
KyeongSoo Yoo 的 Skills 26
ticket-incubate
Use to turn an idea that has no issue/ticket yet into an actionable GitHub issue. Grow a vague notion by divergence, narrow only the gaps via a deficit-based interview, and create the issue once the four required elements (background, goal, done criteria, out-of-scope) are filled. Triggers on "I have an idea", "let's make this a ticket", "create an issue", "interview me", or describing something to make/fix/improve without an issue number, even if the user never says the word "ticket". It is the front of the workbench lifecycle (idea -> ticket-incubate -> issue -> task-start). Distinct from brainstorming/design skills: those work out the how of what to build now; this captures the what of later work as a well-honed ticket.
task-tickets
Use to scan open workbench tickets joined with task branches and lifecycle events into unstarted / claim-attempted / in-progress / awaiting-PR / cleaned-up states. Read-only.
task-submit
Use when the user asks to submit, wrap up, or extract from a task, or when work has reached a point to open a PR.
task-status
Use to see local and remote task status at a glance and diagnose drift (unpushed, stale status, missed cleanup). When you want to know where everything stands before starting or submitting, or after a multi-device handoff.
task-start
Use when the user asks to start a task, pick up an issue, or begin a new attempt, or hands over an issue number to start work.
task-done
Use when the user asks to clean up or discard a task, or to clear the workspace after a PR is merged.
skill-harvest-recommender
Use to retrospect a just-finished (or finishing) session/task and surface procedures repeated by hand that could graduate into a skill, runbook, or utils script. Triggers on "anything from this work worth turning into a skill/automation?", "check what to graduate into tooling", "pull out automation candidates before submit", "let's retro this session", and similar requests that look back at work already done to find repeated patterns. Also fires alongside task-submit's knowledge-harvest gate. It only discovers, routes, and selects candidates; actual skill authoring is delegated to skill-creator/writing-skills and ticket creation to ticket-incubate. Distinct from the authoring stage (you already decided what to build) and from static-scan automation recommenders: this is the retrospective front-end.
docs-query
Use to ask the docs/ wiki a question, run a new task's preflight lookup, or guard right before a decision or approach ("did we already decide / already fail at this?").
docs-lint
Use to check and groom the docs/ wiki: contradictions, staleness, orphan pages, broken cross-references, accumulated supersedes. During maintenance tasks or opportunistically during ingest.
docs-ingest
Use to weave task harvest (decisions/lessons/runbooks/researches) into the workbench docs/ wiki. Called by task-submit, or whenever knowledge that should survive the task needs to land in docs/.
codebase-onboard
Use to onboard (register) an external git repo as a workbench-managed codebase. Triggers on requests like "manage/add this repo in workbench", "register <owner/repo> as a target", "put it in codebases.yaml and clone it", or — without an issue number — "I want to work on this project in workbench". Orchestrates the canonical path (register in codebases.yaml -> `workbench setup` clone -> verify -> name and `role:reference` policy) via a task branch without polluting main. Distinct from /task-start (starting an issue) and `workbench task add-repo` (attaching an already-registered codebase to the current task as a worktree): this is the step before, first bringing a repo into workbench. Plumbing (setup, add-repo) already exists; this only wires the procedure.
caveman
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.
audit-codex-support
Audit this workbench repository for Codex support issues. Use when a task changes or reviews AGENTS.md, README.md, skills/, .agents/skills, .claude/skills, .codex/, .worktrees/, .codebases/, .gitignore, slash-command wording, skill entrypoints, tool state files, stale docs links, or verification commands. Check whether Codex can operate safely and predictably in workbench, including guidance clarity, repo-local skill discovery, search noise, and explicit manual judgment boundaries.
task-start
legacy fixture skill
upgrade-workbench
Use when an existing generated, embedded-engine, staged, or current workbench must be diagnosed, upgraded, migrated, or have its embedded engine removed.
interview-for-personalizing
First step of bootstrapping a new workbench. Through a divergence-first, gap-based interview, draw out the user's persona (language, slug style, required issue elements, PR skeleton, label scheme, harvest fields, gate toggles) and draft it into a temporary .persona/ scratch that generate-workbench consumes. Use this whenever the user wants to set up, create, or personalize a workbench, or talks about "their own rules / conventions" for one — even if they don't say the word "interview". Precedes generate-workbench; together they are the bootstrap flow.
generate-workbench
Second step after interview-for-personalizing. Takes a target repo location or git URL and a .persona/ draft, then scaffolds a minimal workbench (persona-composed AGENTS.md + AGENTS.overlay.md + codebases.yaml + empty docs/) and discards .persona/. The engine (skills+utils) comes from the installed plugin, so the generated repo stays minimal. Use this whenever the user is ready to create their workbench repo from a persona, or asks to generate/scaffold a workbench.
tdd-slice
Use when implementing one selected scenario slice that requires test-first delivery and revision-bound verification evidence.