Claude Code Promts Skills plugin
A plugin with six working skills and a prompt library: routing a task to a prompt, pre-commit checks and skill audits.
- 56 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The skills run shell and Python scripts, and the hooks intercept Bash commands and file writes.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/claude-code-promts-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skills run shell and Python scripts, and the hooks intercept Bash commands and file writes.
This is third-party code. Review the repository files before installing.
What it does
The repository combines a prompt library for Claude and coding agents, built on the APEI cycle (analyze, plan, execute, iterate), with installed skills. The main one, `/find-prompt`, tells you from a task description which prompts to load. The other five run with no model or almost none: `deterministic-checks` finds conflict markers, live-looking keys, debug leftovers and oversized files, `changelog-from-commits` builds a changelog from Conventional Commits, `doc-link-audit` checks links and anchors in docs. Two more skills inspect other people's work. `skill-audit` reviews frontmatter, descriptions and size of skills, while `capability-audit` reads a skill, plugin or MCP server repository before installation without cloning it. As a plugin the repository adds two hooks: one blocks destructive commands and the other stops writing live-looking credentials into files.
Who it is for. Claude Code developers who want ready prompts, fast pre-commit checks and protection from destructive commands.
Good fit when
- Before a commit you need a cheap check for leftover conflicts, keys and debug code.
- A release is going out and the changelog should be built from commits.
- Before installing someone's skill or plugin you want to read its repository without cloning.
- You need to quickly decide which prompts to load for a task such as debugging or a security audit.
Not a fit when
- You need Russian prompts: the repository library is English only.
- Extra hooks are unwelcome: the plugin wires them automatically.
- You do not work in Claude Code: the skills live in `.claude/skills/`.
Example request
/find-prompt diagnose a production failure and check whether it is safe to change this part without a separate auditLimitations
The prompt library is English and the author keeps it so for a consistent tone. Installing the plugin through a marketplace is left to the user because it changes persistent Claude Code state. The two hooks the plugin installs block commands and writes, so in rare cases they can get in the way of a needed operation. The `deterministic-checks` and `doc-link-audit` skills run shell and Python scripts on your machine. `capability-audit` needs GitHub access.
How to disable. Remove the `claude-code-prompts` plugin through the Claude Code plugin manager and the hooks go with it. If you copied skills into `.claude/skills/`, delete their folders.
FAQ
How do I install the plugin?
`/plugin marketplace add Rtur2003/Claude-Code-Promts-Skills`, then `/plugin install claude-code-prompts@claude-code-prompts`. To try it for one session use `claude --plugin-dir` with the path to a clone of the repository.
What does /find-prompt do?
It takes a task description and answers which library prompts to load. The selection rule is to start with one specialist and add another only if the task clearly does not fit.
Which hooks does the plugin install?
A `PreToolUse` hook for Bash commands blocks destructive ones, and for `Write` and `Edit` it stops writing live-looking credentials.
What does deterministic-checks verify?
Conflict markers, live-looking keys, debug leftovers, untracked TODOs and oversized files. The check is a script with no model calls.
Does a skill help with other people's plugins?
`capability-audit` reads a skill, plugin, MCP server or agent toolkit repository on GitHub without cloning and gives evidence before installation.
Discussion
No questions or reviews yet.
Related
PluginsThe author's personal Claude Code toolbox: meta-prompt and todo commands, skills to create skills, hooks and MCP servers, auditors and a Ralph loop
A Claude Code plugin marketplace with 260+ skills and 85 agents, from thinking and writing to company valuation
A plugin of seven skills for Claude Code and Codex: prompts, subagents, skill authoring, editing, web search and a browser.
Open prompt library with a Claude Code plugin, MCP server and CLI: search, fetch and improve prompts and skills from an agent
11 skills for Claude Code: SKILL.md authoring, prompt tuning, refactoring, change review and planning in files.
claude-mods adds 112 skills, 3 agents and 3 commands to Claude Code: stacks, agent operation, defense, session continuity.
README in short
The README shows where to start, lists the six skills with use cases and gives scenario and active prompt tables. Plugin installation and the two protective hooks are described separately.