use-worktree
25Skill Claude CodeCodex
Create one isolated Git worktree for the current task and clean it up after delivery.
not rated 2
changed today A 21 tokens
original MIT
2Stars on the repository
46Mods indexed here, across every type
todayLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown
what-changed
26Skill Codex
A guide for explaining what changed in a previous work session, code difference, or pull request. A pull request is a proposed set of code changes for review before merging.
not rated 2
changed 8d ago A 28 tokens
original MIT
Skill Claude CodeCodex
Explicitly audit and retire clean Git worktrees whose HEAD is durably stored on GitHub, using review-bound approval tokens and .local backups.
not rated 2
changed today A 34 tokens
original MIT
droid-config
28Skill Claude CodeCodex
A review guide for configuring and upgrading Factory Droid, a command-line coding agent, while protecting its credentials and recording a sanitised audit.
not rated 2
changed today A 34 tokens
original MIT
Skill Claude CodeCodex
A configuration tool that keeps prompts and skills in one standard location while linking them into several coding-agent hosts.
not rated 2
changed today B 44 tokens
original MIT
local-test
30Skill Claude CodeCodex
Use only when actually starting, reusing, checking, or stopping a local multi-service integration or project-preview environment, or configuring its lt CLI. Do not use for ordinary code edits, unit tests, static HTML publishing, code review, or remote-only deployment.
not rated 2
changed today A 57 tokens
original MIT
Skill Claude CodeCodex
A visual review workflow for slide-like HTML pages, where each page fills a fixed screen. It checks spacing, margins, empty space, and alignment, then adjusts the layout and produces before-and-after evidence.
not rated 2
changed today A 57 tokens
original MIT
Skill Codex
A required checklist for creating or changing a coding-agent skill. It covers how the skill is triggered, how its instructions and configuration should agree, and what checks to run before delivery.
not rated 2
changed today A 65 tokens
original MIT
Skill Claude CodeCodex
A synchronization workflow that applies a repository's recommended list of user-level skills as symbolic links on a computer. A symbolic link is a filesystem shortcut pointing to another file or folder.
not rated 2
changed today A 46 tokens
original MIT
skill-test
34Skill Claude CodeCodex
A workflow for testing a coding-agent skill in an isolated run with a fresh subagent. It compares the result with predefined success criteria and can use blind A/B tests to evaluate behavior.
not rated 2
changed today A 24 tokens
original MIT
Skill Claude CodeCodex
A review workflow for finding internal instructions or leftover prompts that are visible in a website or document. It distinguishes text found in files from text visible in a running website.
not rated 2
changed today A 60 tokens
original MIT
submit-pr-mr
36Skill Claude CodeCodex
A workflow for writing a pull request or merge request description. A pull request or merge request is a proposal to merge code changes into a shared repository.
not rated 2
changed today A 61 tokens
original MIT
preflight
37Skill Claude CodeCodex
Verify before implementation starts that a spec or task instruction can actually be completed in this environment: credentials, permissions, tools, and external dependencies are checked and dispositioned.
not rated 2
changed today A 36 tokens
original MIT
Skill Claude CodeCodex
Declare a repository's lifecycle stage (experimental or production) in its agent instruction file so agents stop adding compatibility layers that no consumer needs. Use only when the user explicitly invokes $project-stage, typically once when starting a project, once more when it graduates to production, or when…
not rated 2
changed today A 66 tokens
original MIT
Skill Claude CodeCodex
A tool that fixes failed checks from the latest local agent-readiness report. It examines the related repository areas, makes focused improvements, and verifies the changes; it runs only when explicitly invoked.
not rated 2
changed today A 41 tokens
original MIT
Skill Claude CodeCodex
A read-only audit of a Git repository that produces a local report about how ready the codebase is for autonomous coding agents. It scans the repository and saves a history of the report without sending it to a remote service.
not rated 2
changed today A 43 tokens
original MIT
Skill Claude CodeCodex
A task-dispatching skill that sends explicitly requested work to another configured coding agent or thread. It chooses a difficulty and task type, while using the user's configured tools and models.
not rated 2
changed today A 45 tokens
original MIT
html-preview
42Skill Codex
Use only when an agent-generated static HTML report, chart, or demo must be published as a browser link for the user, or when renewing, inspecting, or deleting an existing HTML preview. Do not use for merely writing HTML files, editing application code, running project servers, or publishing production sites.
not rated 2 today A 63 tokens
original MIT
anti-ai-slop
43Skill Claude CodeCodex
A review skill for finding unnecessary AI-generated clutter in code and documentation. It looks for repeated explanations, leftover prompt text, defensive permission wording, unwanted abstractions, and duplicate implementations.
not rated 2
changed today A 54 tokens
original MIT
Skill Claude CodeCodex
Upload local screenshots, videos, or diagnostic attachments as GitHub Release Assets, embed them in an issue, PR body, or comment, and read the published result back online. Flow skill: explicit user invocation, or reuse by an authorized caller workflow.
not rated 2
changed today A 59 tokens
original MIT
Skill Claude CodeCodex
Maintain a non-maintainer public fork through its local aggregate workspace.
not rated 2
changed today A 20 tokens
original MIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: