liby

12 mods across 1 repository, 149 stars between them.

dotfiles AGENTS.md

01

liby/dotfiles

Instructions file CodexOpenCode

AGENTS.md instructions for liby/dotfiles, covering workflow, encrypted files and project invariants.

149 2d ago A 3,204 tokens

commit

02

liby/dotfiles

Skill Claude CodeCodex

Create or amend a local Git commit from the relevant changes. Use for every commit you are about to create, regardless of how the request arrived, which repository it targets, or what tooling wraps git. Stage only in-scope files and write a repository-matching message. Not for message drafts or branch creation.

149 2d ago A 64 tokens

deslop

03

liby/dotfiles

Skill Claude CodeCodex

Clean changed code, tests, comments, and documentation before final validation: reuse established mechanisms, reduce accidental complexity and cost, remove AI artifacts, and verify type-driven refactors against runtime evidence. Use when the user invokes /deslop, asks to run deslop, or governing instructions require…

149 2d ago A 70 tokens

gh

04

liby/dotfiles

Skill Claude CodeCodex

Operate GitHub through the gh CLI for GitHub issues, pull requests, repos, workflow data, comments, and GitHub-hosted Agent Skills. Use when the user gives a GitHub URL, owner/repo#123, asks about a GitHub issue/PR/workflow, or asks to preview/install/update a skill from GitHub. Not for GitLab URLs or local skill…

149 2d ago A 89 tokens

glab

05

liby/dotfiles

Skill Claude CodeCodex

Operate GitLab through the glab CLI for GitLab merge requests, issues, pipelines, discussions, repos, and MR descriptions. Use when the user gives a GitLab URL, asks about a GitLab MR/issue/pipeline, mentions glab, or asks to draft/update an MR description. Not for GitHub URLs or purely local git tasks that do not…

149 2d ago A 94 tokens

herdr

06

liby/dotfiles

Skill Claude CodeCodex

Control Herdr, a terminal multiplexer for coding agents. Use only when the user explicitly mentions Herdr or asks to use Herdr to inspect or control panes, tabs, workspaces, commands, or another agent. Do not use merely because a task could benefit from a background terminal, delegation, or parallel work. Requires…

149 2d ago A 74 tokens

new-branch

07

liby/dotfiles

Skill Claude CodeCodex

Create a new Git branch for a task using the repository's naming convention. Use when the user asks to create a branch, or a requested task such as opening an MR or PR needs one as a step. Not for switching to an existing branch.

149 2d ago A 54 tokens

oracle

08

liby/dotfiles

Skill Claude CodeCodex

Use Oracle only when the user explicitly requests Oracle, ChatGPT Pro, ChatGPT Deep Research, a ChatGPT Project consultation, continuation of an Oracle session, or Oracle API mode. Not for requests limited to other consultants, ordinary review or research, or inspection of existing ChatGPT content.

149 2d ago A 59 tokens

review

09

liby/dotfiles

Skill Claude CodeCodex

Review remote merge requests or pull requests, branches, commits, diffs, and uncommitted code changes for correctness and regression risk, reporting only evidence-backed findings. Use for pre-merge code review, review findings, and MR/PR URLs. Not for prose review, skill-authoring audits, or implementation requests.

149 2d ago A 66 tokens

set-goal

10

liby/dotfiles

Skill Claude CodeCodex

Create and start an outcome-based, verifiable goal from an explicitly requested task. Use for /set-goal, requests to set or start a goal, or long-running goal mode. Not for goal-setting discussion, ordinary planning, continuing an active goal, or direct /goal Read ... handoffs.

149 2d ago A 65 tokens

snow

11

liby/dotfiles

Skill Claude CodeCodex

Run read-only Snowflake inspection and SELECT queries with the installed Snowflake CLI. Use whenever a task needs live Snowflake data. Not for dbt compilation that needs no live warehouse access.

149 2d ago A 40 tokens

write-skill

12

liby/dotfiles

Skill Claude CodeCodex

Create, test, improve, or audit agent skills and reusable agent instructions. Use for evidence-driven refinement from feedback or failed runs, skill routing, workflow or wording optimization, validation, splitting or merging skills, and instruction-owner audits. Not for application-code review, generic documentation…

149 2d ago A 79 tokens