What it does
This repo is a collection of 18 skills
- Connects Instruction files the AI reads, with steps that call an outside service
- Account or key Some steps need a key for Exa, CONTEXT7_API_KEY, GitHub (per the README)
- Category Automation and agents
Maker’s original description
Agent toolkit for OpenCode, Codex, and DeepSeek Harness: skills, MCP, background agents, and durable job trackingFor makers: add a badge to your README
If you made this, paste this code into your README. The badge links back to this page.
[](https://mcp.popnx.com/en/skills/omagents-omagents)Similar Skills
Same kind of work, ordered like the Skills list
18 skills (instruction sets) by omagents that an AI reads and follows · not an MCP server
Pick the skill to install
Install only the ones you use. Open a skill to see how to install it
Install the whole set (18)
If you want every one, let the AI install them in one go. Know 2 things first
- Every installed skill’s description is loaded into every chat. The more you install, the easier it is for the AI to pick the wrong one
- Some skills duplicate tools the AI already has built in. Installing only the ones you use works better
Uploading to Claude on the web and apps: you need 18 uploads, one ZIP per skill. Open each skill in the list below to download it
Checked against official docs Nov 12, 2026 · Source: support.claude.com, code.claude.com, learn.chatgpt.com, cursor.com
- agents-python-tools
Route agent Python tooling to a dedicated venv at ~/.venvs/omagents, separate from any project's Python environment. Use whenever the agent needs to install or run a Python tool/library to complete a task — e.g. openpyxl to generate Excel, pdf2md/pdfplumber to convert PDFs to Markdown, Pillow to ge…
- ast-grep
AST-aware code search and rewrite across 25+ languages using the ast-grep (sg) CLI. Trigger when the agent needs structural code matching, pattern-based search, or pattern-based rewriting. Falls back to grep if ast-grep is not installed.
- deep-research
Run a multi-source, iterative deep research workflow using structured items × fields and produce a Markdown report.
- doctor
Diagnose OmAgents installation and configuration. Trigger when the user says 'doctor', 'diagnose', or encounters issues with MCP servers, skills, venv, or plugin loading. Checks plugin registration, MCP connectivity, Python venv, skill discovery, and PATH injection.
- github-triage
Triage GitHub issues using a durable loop workflow. Each issue is classified, labeled, and recorded before moving to the next. Trigger when the user says 'triage issues', 'classify issues', 'organize issues'.
- hyperplan
Adversarial plan review with 3 parallel critics using loop engine tracking. Each critic attacks the plan from a different angle: security, architecture, and edge cases. Trigger when the user says 'hyperplan', 'review plan', 'critique plan', or wants adversarial analysis of a design.
- init-deep
Auto-generate hierarchical AGENTS.md files throughout a project. Trigger when the user says 'init-deep', 'generate AGENTS.md', or wants to set up project context for AI agents. Scans the codebase and creates context files at each meaningful directory level.
- lsp-guide
Guide agents to use the right code intelligence tool: OpenCode's native LSP, codegraph MCP, grep, or ast-grep. Trigger when the agent needs code navigation, definition lookup, reference finding, renaming, or diagnostics.
- markitdown-converter
Use Microsoft MarkItDown to convert documents and files to Markdown for text extraction and parsing. Trigger this skill whenever the task involves: reading/extracting text from PDF, DOCX, PPTX, XLSX, XLS, HTML, CSV, EPUB, IPYNB, images (with OCR/LLM), audio (with transcription), Outlook MSG, RSS/AT…
- officecli
Create, analyze, proofread, and modify Office documents (.docx, .xlsx, .pptx) using the officecli CLI tool. Use when the user wants to create, inspect, check formatting, find issues, add charts, or modify Office documents.
- parallel-execution
The orchestrator agent automatically parallelizes work using background tasks. Use /ps to check status, cancel_task to cancel.
- playwright-web-scraping
Use Playwright (Python) for all web scraping and page-fetching tasks instead of curl, wget, or requests. Trigger this skill whenever the task involves: fetching a web page's content, scraping data from a URL, reading a website's text, extracting information from HTML, or any task where a web page n…
- pre-publish-review
Pre-publish release gate using a durable loop workflow. Checks version, changelog, tests, lint, Python scripts, git status, and docs before publishing. Trigger when the user says 'pre-publish', 'release check', 'ready to publish', or before running npm publish.
- refactor
Systematic code refactoring using a durable loop workflow. Each refactoring target is processed, verified, and marked complete before moving to the next. Trigger when the user says 'refactor', 'restructure code', or wants to improve code without changing behavior.
- remove-ai-slops
Clean up AI-generated code artifacts using a durable loop workflow. Trigger when the user says 'clean AI code', 'remove slops', 'remove AI artifacts', or notices low-quality AI-generated code. Processes files one by one with verification.
- remove-deadcode
Find and remove unreferenced code using a durable loop workflow. Each candidate is verified, removed, tested, and marked complete before moving to the next. Trigger when the user says 'remove dead code', 'clean unused code', 'find unused'.
- tech-debt-audit
Systematically audit a codebase for technical debt using a durable loop workflow. Each audit category is scanned, findings collected, and recorded before moving to the next. Trigger when the user says 'tech debt audit', 'audit code', 'code health check'.
- work-with-pr
Manage the full PR lifecycle using the github MCP: create, update, review, merge. Trigger when the user says 'create PR', 'update PR', 'review PR', 'merge PR', or needs to work with pull requests.
How a Skill differs from MCP
MCP = the hands
Connects the AI to an outside service or data, such as reading Notion or creating an invoice. A program or server has to be running.
Skill = the manual
A text file that teaches the AI how to do a task. It connects to nothing.
Some overlap: a few Skills run scripts that call tools, and a few MCP servers only send knowledge. Start from the task you want done.
Works with
CodexGitHub CopilotOpenCode
From the maker’s README and the folders in the repo
Install
Pick a skill first
This repo has 18 skills. Pick one from the list above; its page shows every way to install it, including letting the AI install it and a ZIP
Discussion
Ask, share a tip or report that it does not work. Sorted by what people marked helpful.
No posts yet. Be the first to ask or share.
Posts here are written by users, not the PopNx team.
From GitHub
- GitHub stars
- 359
Stars are for the whole repo, not each skill - Maker
- omagents
- License
- MIT
- Last code change
- Nov 16, 2026
- SKILL.md files in the repo (including translated copies and templates)
- 18
- Topics
- agent-skills, ai-agents, codex, deepseek-harness, dsh, dsh-plugin, mcp, opencode