Terminal Skills
Over 1,100 ready-made skills for Claude Code, Codex, Gemini CLI and Cursor, plus an MCP server that picks a skill for the task
- 192 ★ on the repository
- Last commit
- Apache-2.0
Extended access
Why this level
- Installed skills instruct the agent to run commands and call network services.
- The hosted MCP servers receive an account token, and SEOG manages company profiles and replies to reviews on the user's behalf.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/terminal-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.
- Installed skills instruct the agent to run commands and call network services.
- The hosted MCP servers receive an account token, and SEOG manages company profiles and replies to reviews on the user's behalf.
This is third-party code. Review the repository files before installing.
What it does
Terminal Skills is an open library of ready-made `SKILL.md` files following the Agent Skills standard. The `skills` folder holds over 1,100 skills: PDF and Excel processing, code review, test generation, Docker, CI/CD, SQL tuning, scraping, SEO and email campaigns. Each skill is a separate file. It is installed with `npx terminal-skills install` into a Claude Code project or placed by hand in the Codex, Gemini CLI and Cursor directories. Two skills rely on hosted MCP servers. The Terminal Skills server finds a skill for a task with the `search_skills`, `get_skill` and `plan_task` tools, so the agent locates and loads the right one itself. The SEOG server offers about 90 tools for local SEO: the Google company profile, map ranking checks, review replies, site audits and PDF reports. The `use-cases` folder next to it holds hundreds of step-by-step walkthroughs of common tasks.
Who it is for. Developers and agent operators who want a large ready-made set of skills for common tasks instead of writing their own from scratch.
Good fit when
- An agent needs a skill for a routine task: PDF parsing, code review, tests or Docker setup.
- The agent picks a suitable skill itself from a task description through the Terminal Skills MCP server.
- Local SEO for clients is run from the agent: map rankings, review replies and reports through the seog skill.
- A repeated workflow is packed into a skill following the agent-workflow-packager example.
- One set of skills is handed to several agents by copying files into their directories with `curl`.
Not a fit when
- You need one vetted skill from a specific vendor: a general library does not replace official sets.
- A skill would be installed unread: instructions in the library give the agent access to commands and the network.
- You work offline: installation through `curl` and the hosted MCP servers need the internet.
Example request
Install the pdf-analyzer skill from Terminal Skills and extract the tables from `report.pdf` into CSV.Limitations
Skills sit in a flat list under `skills` and the repository has no overall quality rating, so it is worth opening `SKILL.md` before installing to see which commands and services it calls. The `npx terminal-skills install` command is documented for Claude Code, while for Codex, Gemini CLI and Cursor manual `curl` commands are given with the `.codex/skills`, `.gemini/skills` and `.cursor/skills` directories. The hosted MCP servers need an account token or an OAuth sign-in in Claude, and SEOG also needs an account at seog.ai. Some skills have their own dependencies, such as Python libraries and binaries, listed in the `compatibility` field at the top of the file.
How to disable. Delete the skill file from the project's `.claude/skills` directory (or from another agent's directory). Remove the MCP entry from the agent configuration and revoke the token in your account.
FAQ
How do I install a Terminal Skills skill in Claude Code?
Run `npx terminal-skills install pdf-analyzer`, replacing the name with the one you need. The same file can be fetched by hand: `curl -sL https://raw.githubusercontent.com/terminal-skills/skills/main/skills/pdf-analyzer/SKILL.md -o .claude/skills/pdf-analyzer.md`.
Does the library work with Codex, Gemini CLI and Cursor?
Yes, through manual copying. The same `curl` command puts `SKILL.md` into `.codex/skills`, `.gemini/skills` or `.cursor/skills`. Automatic installation with `npx` is documented only for Claude Code.
What is the Terminal Skills MCP server for?
It matches a skill to the task: the `search_skills`, `get_skill` and `plan_task` tools search the whole catalog and load the needed file. Connect `https://api.terminalskills.io/api/mcp` as a connector in Claude over OAuth, or in a CLI with a token in the `Authorization` header.
How do I connect SEOG?
Point the client at `https://api.seog.ai/mcp` and use a token from your seog.ai account. If you only want the instructions without the tools, install the skill separately with `npx terminal-skills install seog`.
How many skills are there and where can I browse them?
The `skills` folder holds over 1,100 `SKILL.md` files. The full catalog is open on terminalskills.io, and the README gives a table of 25 examples with categories.
Discussion
No questions or reviews yet.
Related
SkillsThe author's personal set: shared agent rules and dozens of skills for GitHub, the browser, macOS, messaging and releases, plus a sync script
A set of independent skills against drift in large codebases: hard-cut, root-cause tracing, lean testing and CI right-sizing.
skills-npm links agent skills shipped in your dependencies in node_modules and listed in the skills field of package.json
110 skills for Claude Code and Codex and seven Claude Code agents: development, product, UI, diagnostics and operations, installed whole or as a bundle
61 small skills for Codex across 13 areas: codebase orientation, tests, migrations, security, CI and pull request review
CLI and MCP tools for finding and installing skills, with SKILL.md linting and flags for network use, scripts and secrets
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| Terminal Skills (this card) | no | 192 | |
| Anthropic skills | no | 188,047 | Anthropic's official examples covering documents, design and the Agent Skills spec; Terminal Skills collects over a thousand skills for everyday development and business tasks |
| Agent Skills by Addy Osmani | no | 113,521 | 25 engineering skills along the development lifecycle; Terminal Skills is broader in topics and adds a hosted routing MCP server |
| Awesome Claude Skills by Composio | no | 78,824 | a directory linking to other people's skills and plugins; Terminal Skills keeps the skill files themselves in one repository |
README in short
The README briefly explains installation for four agents, how to connect the two MCP servers and shows a table of 25 examples by category. The rest of the catalog lives on the website and in the `skills` and `use-cases` folders.
README badge
Original SKILL.md
--- name: pdf-analyzer description: >- Extract text, tables, metadata, and structured data from PDF files. Use when a user asks to read a PDF, parse a PDF, extract data from a PDF, summarize a PDF document, pull tables from a PDF, or convert PDF content to structured formats like JSON or CSV. Handles single and multi-page documents, flags scanned PDFs for OCR, and reads borderless and password-protected PDFs. license: Apache-2.0 compatibility: "Python 3.10+; pip install pdfplumber (required) and pymupdf (optional, fallback and speed); OCR needs the Tesseract binary plus pytesseract" metadata: author: terminal-skills version: "1.1.0" category: documents tags: "pdf, document, extraction, tables, parsing" --- # PDF Analyzer ## Overview Extract text, tables, and structured data from PDF files and convert them into usable formats. This skill handles text extraction, table detection, metadata reading, and output formatting for single or multi-page PDFs. ## Instructions When a user asks you to analyze, read, parse, or extract data from a PDF file, follow these steps: ### Step 1: Identify the PDF and goal Determine the file path and what the user wants extracted: - **Full text**: All readable text from every page - **Tables**: Structured tabular data - **Metadata**: Title, author, creation date, page count - **Specific sections**: Targeted content from certain pages - **Summary**: A condensed version of the document contents ### Step 2: Choose the extraction method Write a Python script using pdfplumber (prefer it for tables) or PyMuPDF (`import pymupdf`; the old `fitz` name is deprecated and only warns). Pass `password=` for encrypted files; pdfplumber raises an empty `PdfminerException` otherwise. **For text extraction:** ```python import pdfplumber