Awesome Skills
Twenty skills for reading unfamiliar code, reviewing documents and commits, translating Markdown and OpenSpec-driven development
- 92 ★ on the repository
- Last commit
Extended access
Why this level
- Some skills change files and git history, and `web-content-downloader` sends page addresses to the external Jina Reader service.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/forceinjection-awesome-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.
- Some skills change files and git history, and `web-content-downloader` sends page addresses to the external Jina Reader service.
This is third-party code. Review the repository files before installing.
What it does
Awesome Skills is a collection of twenty skills by the author behind 原力注入. The main one, `code-reader`, reads an unfamiliar codebase with three roles (a technical writer, a QA engineer and a junior developer) and checks understanding with a closed-book exam, producing a `SKILL.md` for each module. `project-analyzer` builds on it to write a seven-chapter architecture report on a third-party repository, with code and execution flow taking most of the space. The other skills cover everyday text and repository work. `doc-reviewer` reviews documents in four types (structure, content, links, formatting), `md-summarizer` and `md-translator` work with Markdown, and `md-link-checker` checks links with a cache and parallel requests. `update-submitter` groups changes from `git status` and `git diff` and prepares Conventional Commits, while `self-review` checks the diff before a commit. `openspec-assistant` runs specification-driven development with three roles. There are also `humanizer-zh` for cleaning Chinese text of AI traits, `drawio-designer`, `pptx-reader`, `pptx-editor`, `ontology` and `reference-organizer`. Texts meant for the agent are in English, results for people are in Chinese.
Who it is for. Chinese-speaking developers and technical writers who study other people's repositories, write documentation and keep working notes in Markdown.
Good fit when
- You need to understand an unfamiliar repository and want a `SKILL.md` for each module for later sessions
- A third-party project needs an architecture report with a walkthrough of its execution flow
- Before a push you want to check the diff against a list of typical mistakes and get verdicts of fixed, pending and not confirmed
- A documentation folder has accumulated broken links and stale paths that have to be found in one pass
- You need a Draw.io diagram with AWS icons exported to PNG with a transparent background
Not a fit when
- You need one skill for a narrow task: all twenty at once is overkill
- You need results in Russian or English: the reports and skill output are tuned for Chinese
- The task is unrelated to code and documents: the skills cover code reading, Markdown, presentations and diagrams
Example request
/code-reader ./vendor/some-lib ./skills-out: study this library and prepare a SKILL.md for each module.Limitations
The README has no install command: skills sit in the `skills/` folder and are copied into the skills directory of your agent. Reports for humans come out in Chinese, and `humanizer-zh` targets Chinese text. `web-content-downloader` fetches page text through Jina Reader, so addresses go to an external service. `pptx-reader` and `drawio-designer` rely on LibreOffice, Poppler and headless Chrome, which must be installed on the system. The `dir-organizer` and `update-submitter` skills change files and git history but show a plan and wait for consent first. The `ontology` skill is launched by a Python script and not by an agent command.
How to disable. Delete the skill folders from the agent skills directory. Generated reports and `SKILL.md` files stay in the project and are removed by hand.
FAQ
What does the `code-reader` skill do?
It reads a codebase with three roles (a technical writer, a QA engineer and a junior developer) and checks understanding with a closed-book exam. The result is `SKILL.md` files per module that any agent can load later.
Why produce `SKILL.md` files instead of separate agents?
The author argues that an agent per module multiplies roles and hard-codes logic into prompts. A knowledge file can be loaded by any agent when needed.
How do I install the skills?
The README has no command: copy the folders from `skills/` into the skills directory of your agent. Triggers such as `/code-reader` and `/doc-reviewer` are listed in the table.
Are external programs needed?
`pptx-reader`, `pptx-editor` and `drawio-designer` need LibreOffice, Poppler and headless Chrome. `web-content-downloader` calls Jina Reader.
Which language do the skills use?
`SKILL.md` files are in English for model accuracy and final reports for people are in Chinese. The exceptions are `dir-organizer` and `doc-reviewer`, written in Chinese.
Discussion
No questions or reviews yet.
Related
SkillsSimplified Chinese localization of Addy Osmani Agent Skills: 25 engineering skills and 9 lifecycle commands for several agents.
Five skills for daily development: requirement interviews, AGENTS.md rules, PR review triage and single-file HTML design.
Dimillian Skills
Skills
A set of 16 engineering skills for agents: Apple platform work, macOS packaging, React performance, code review and refactoring
kyy-logs code-review-skill
code-review-skill
A code review skill with four independent probes, evidence-backed findings and a merge gate derived from JSON
my-skills by akitaonrails
akitaonrails my-skills
Fabio Akita's personal set of 24 skills for Claude Code, Codex and opencode: releases, PR audits, security and fact-checking
A skill set against AI slop: UI and typography audits, change review, releases, docs and SEO.
README in short
The README lists twenty skills with triggers, the language principles (English for the agent, Chinese for people), the reason for choosing `SKILL.md` over agents and advice on skill structure and descriptions.