RashadAnsari/myagents

**One command to give every AI coding agent you use the same memory, reviewers, and working rules.**
Coding agents start every session from zero. They forget what your team decided last week, skip the conventions in AGENTS.md, guess at library versions, and review code in one generic pass. myagents fixes that across Claude Code, Cursor, and Codex at once:
- **Memory that persists.** Agents look up project decisions and your preferences before they start, and save new ones when they finish. Memory follows the repository across clones and machines.
- **14 specialist reviewers.** Security, performance, architecture, tests, accessibility, database, dead code, and more. Run all of them in parallel, pick one, or point them at a GitHub pull request.
- **Rules that stay on.** Research before guessing, the simplest change that works, no filler in answers. Hooks load them at the start of every session and repeat them on every prompt, so they hold up in long conversations.
- **One install, three tools.** The installer finds the tools you have and sets up each one the way that tool expects. Rerun it to update.
---
Quick start
On macOS, run this. It installs any missing tools it needs with Homebrew, and installs Homebrew first if you don't have it:
Then try it in any project:
To update, run the same install command again.
---
Architecture
The diagram is generated with archify from docs/architecture.json. Run make diagram to render and export it again.
---
Contents
- Architecture
- Plugins: albino, archify, context7, emil-skills, ponytail, security-audit, ui-ux-pro-max
- Platform support
- Agent memory
- Installing, updating, and uninstalling
- Contributing
- License
---
Plugins
albino
The main plugin: review agents, skills, MCP servers, and hooks.
Workflow skills
Skills you run by name. They replaced the old slash commands, which all three tools are moving away from.
| Skill | What it does |
|---------|--------------|
| /reviewcrew | Runs every review agent in parallel and writes a consolidated REVIEW_REPORT.md |
| /reportloop | Walks through REVIEW_REPORT.md one issue at a time, by severity: explains it, then fixes or skips it |
| /audit [reviewer] [path] | Runs one reviewer of your choice, optionally on one path, and prints its findings |
| /pr-review | Reviews a GitHub pull request: picks the reviewers that fit the changed files, loads memory, checks ripple effects, lets you choose which findings to post, and submits the review with gh |
| /commit | Stages all changes and writes a commit message in the project's existing style |
| /commit-pr | Commits and opens a pull request, creating a branch first when you are on main or master |
| /remember | Saves something to project or user memory, or this session's learnings when given nothing |
| /forget | Finds and removes matching entries from project and user memory |
| /enrich | Mines the 100 most recently merged pull requests for decisions, conventions, and gotchas, and saves them to project memory |
| /statusline | Sets the Claude Code status line: folder, git branch and dirty state, model, context usage, and time |
Run one with / in Claude Code (or /albino: when another plugin uses the same name) and Cursor, and with $ or /skills in Codex. Claude Code fills in the arguments after the name; Cursor and Codex do not pass arguments to skills, so give the subject in the same message instead. commit, commit-pr, pr-review, forget, enrich, reviewcrew, and statusline run only when you type them, never on the agent's own initiative.
Review agents
Used by /reviewcrew, /audit, and /pr-review, or on their own.
| Agent | Looks for |
|-------|-----------|
| security-reviewer | Injection, XSS, auth flaws, SSRF, JWT misuse, and supply chain risk |
| code-reviewer | Correctness, anti-patterns, concurrency, and memory bugs |
| architecture-reviewer | Structure, coupling, cohesion, SOLID, and duplication |
| performance-reviewer | Bottlenecks, complexity, slow queries, caching, and scalability |