liaoann/skills

1

diagnose-bug Diagnose a reported product, runtime, or performance bug before fixing it. Use when the user reports broken, slow, su…

liaoann/skills

45

2

ask-me Router for my process skills. Type /ask-me when unsure which workflow fits the situation.

liaoann/skills

42

3

implement-plan Implement an agreed code-change plan. Use when a plan exists or the user asks the agent to execute planned work, edit…

liaoann/skills

42

4

review-change Review a completed or in-progress code change from a fresh reviewer perspective. Use when the user asks for code revi…

liaoann/skills

42

5

study-repo Use for repo/package questions, feature tracing, usage checks, and dependency uncertainty about behavior, contracts, …

liaoann/skills

42

6

architecture Module-boundary and dependency-hygiene principles for organizing a codebase. Use when placing new files, deciding whe…

liaoann/skills

41

7

engineering-quality Maintainable-code principles for clarity, comments, error context, small local refactors, and algorithmic shape. Use …

liaoann/skills

41

8

tdd Test-driven development via red-green-refactor. Use when the user wants to build a feature or fix a bug test-first, m…

liaoann/skills

41

9

plan-it Plan a code change before implementation. Use when the user wants to add, modify, remove, or fix behavior and needs a…

liaoann/skills

40

10

agentic-change-governance Agent authority and scope-control principles. Use when deciding whether an agent is authorized to make or propose a c…

liaoann/skills

39

11

ci-triage Triage and boundedly remediate failing CI or local validation checks. Use when builds, type checks, lint, tests, GitH…

liaoann/skills

39

12

codebase-stewardship Codebase-coherence principles for fitting a change into existing patterns. Use when choosing names, file placement, e…

liaoann/skills

39

13

contracts Contract-design principles for APIs, schemas, events, jobs, module exports, configuration, and other cross-boundary i…

liaoann/skills

39

14

property-based-testing Use when generated inputs can search for unknown counterexamples to a stated behavioral property, law, invariant, rou…

liaoann/skills

39

15

reviewable-change Diff hygiene principles for human-auditable changes. Use when deciding how to split, limit, or report a diff so behav…

liaoann/skills

39

16

persistent-side-effects Persistent side-effect approval contract. Use before creating files or directories, staging or committing changes, cr…

liaoann/skills

36

17

open-pr Write a PR description and open the pull request. Use when the user asks to open a PR, write a PR description, or pre…

liaoann/skills

22

18

drizzle-orm Use when working with Drizzle ORM on any SQL driver (Cloudflare D1, Postgres, MySQL, SQLite, LibSQL) — writing or edi…

liaoann/skills

18

19

codex-session-continuity Use when the running agent's model is Codex or the GPT-5 family and any workflow phase finishes — a plan from /plan-i…

liaoann/skills

11

20

devcontainer-exec Use when a project is developed inside a Dev Container — the repo has a `.devcontainer/` directory, or CLAUDE.md/AGEN…

liaoann/skills

10

21

knip Use when configuring or debugging Knip — creating or editing a knip config, tuning entry/project globs, silencing fal…

liaoann/skills

7

22

agent-instruction-files Placement rules for agent instruction files (AGENTS.md, CLAUDE.md, .cursorrules, .github/copilot-instructions.md) and…

liaoann/skills

6

23

plan-change Plan a code change before implementation. Use when the user wants to add, modify, remove, or fix behavior and needs a…

liaoann/skills

2

24

codex-search Fallback web research through the locally authenticated Codex CLI. Use only when the task requires searching the exte…

liaoann/skills

1