x
v0.11.99Dima's own skills — code core values + language guides, CST session handoff, commit workflow, RL task authoring, prompt review, agentic loops, pm ticket quick-ops, cross-repo conventions
By Dima VakatsiienkoLicense: MIT0 GitHub starsUpdated 14 hours ago
Directory evidence
- Runtimes
- Claude Code
- Parsed components
- 25 skill or MCP entries
- Source updated
- Sep 23, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install x for Claude Code
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install x@agent-plugin-marketplacePaste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/dvakatsiienko/frameClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is home/.claude/plugin-x/.
Plugin files
├── .claude-plugin/plugin.json├── skills/browser-headless/SKILL.md├── skills/cmt/SKILL.md├── skills/coder-brief/SKILL.md├── skills/github-contrib/SKILL.md├── skills/guide-code/SKILL.md├── skills/guide-conventions/SKILL.md├── skills/guide-react/SKILL.md├── skills/guide-typescript/SKILL.md├── skills/guide-ui-ux/SKILL.md├── skills/handoff/SKILL.md├── skills/handoff-ingest/SKILL.md├── skills/humanize/SKILL.md├── skills/humanize-audit/SKILL.md├── skills/mobile-mode/SKILL.md├── skills/notes/SKILL.md├── skills/pm/SKILL.md├── skills/pre/SKILL.md├── skills/queue/SKILL.md├── skills/remind/SKILL.md├── skills/step-by-step/SKILL.md├── skills/sweep-issues/SKILL.md├── skills/verifier-brief/SKILL.md├── skills/walkthrough/SKILL.md├── skills/writing-for-humans/SKILL.md└── skills/yt-transcript/SKILL.md
Included Skills25
Load BEFORE any browser check of a web app, in every session — cli-born or Code-tab with the Browser pane — «verify in a browser», «does it render», «screenshot», «hover», «check the console», «test at mobile width», a chart that renders blank in the pane — and before writing any playwright or puppeteer script.
«/cmt», «commit», «slay» — load on sight, even mid-sentence or after another instruction («approve, /cmt y, and next», «ask coder to sync, then commit and slay», «you can commit them too», «vibe», «mana»). Load EVERY time a git commit is about to be created, in any repo.
the coder contract — typed by dima into a fresh coder session as `/x:coder-brief <BYT-N|DOT-N> [job]`, or pasted by cclio into a `--bg` spawn prompt. never auto-loaded.
Load BEFORE any `gh pr` / `gh issue` command, before writing a PR title or description, before filing or editing an issue, and whenever asked to watch or babysit a PR.
Load EVERY time you are about to write, edit, refactor or review source code — a .ts/.tsx/.swift/.sh/.py file, a function, a test, a script, a hook — BEFORE the language-specific guides. Not for settings dialogs, hotkeys, snippets, raycast/app configuration, tickets or prose.
Load BEFORE designing a shape that is expensive to change later — «new route», «url shape», «api surface», «where should this file live», «package.json shape», «rename the family» — and whenever a review comment sounds like "we don't do it that way here". Not for a hotkey, a setting, or a tool pick.
Load EVERY time you write, edit, or review React/JSX components — always paired with guide-typescript for the type side.
Load EVERY time you write, edit, or review TypeScript in any repo — .tsx work loads it alongside guide-react.
Load EVERY time you render anything a human looks at — html, react/jsx, an artifact page, a chart, a tui — before the first element is written or reviewed.
Load on /handoff with any argument shape — focus, spawn, <session-id|name> push, list, peek <slug>, delete — and on an incoming HANDOFF REQUEST message.
Load on /handoff-ingest or "grab/pull/ingest handoff" — session id/name to request from a live CC peer; bare, topic keyword, or "cw" for the shared store.
Use whenever the user asks to "humanize", "make this sound more human", "rewrite to avoid AI detection", "make this less AI-sounding", "add a human voice", or "write like a person". Also use when the user pastes text and asks why it reads as robotic, generic, flat, or AI-like, or when generating new text in a register where AI tells (em dashes, semicolons, hedges, banned vocabulary like "delve", "leverage", "robust") would damage credibility. Routing: this skill REWRITES existing text; drafting a NEW message for a human starts in writing-for-humans, which calls this one for repairs.
Use when someone asks "does this sound AI?", "check if this is AI-written", "what gives this away as AI", "run ai-check on this", or "score this text". Also use when reviewing a draft for AI tells before publishing, or when a piece of text reads as suspiciously polished, generic, or pattern-y and the user wants a forensic breakdown of why. Routing: this skill only SCORES text; drafting starts in writing-for-humans, rewriting in humanize.
Dima types /mobile-mode from ipad/phone — flips the session into mobile mode (short, one step at a time, finger-cheap confirms) until he says he is back at the mac.
Load BEFORE any read or edit of an obsidian vault note or a notion page — «inbox», «flowlog», «the vault», «obsidian», «notion», «all notes», «edit the note», «append to inbox», «rename the note», «update the notion page», a vault path, a notion url.
Load BEFORE any `linear` command runs — reads included — any FRM-N/BYT-N mention, "read/check <ticket>", create/update/comment/close/triage, "save this idea as a ticket", "ticketify", scope changes to tracked work, or /x:pm.
Load when Dima types /pre or marks an item "pre:", or asks for a tiny thing done right now before the main thread continues.
Load when Dima types /queue, or says to do something after the current thing / later in the session.
Load when Dima types /remind, or asks to be reminded of something / to drop or list a reminder.
Load when Dima types /step-by-step <what>, or asks to go through a set one by one / file by file — when how to batch-resolve is unclear and each item needs his steer.
Invoke for an occasional deep-clean of shipping, inherited, or vibe-coded work (30-60 min, heavy tokens) — NOT everyday review; "review only" in the argument skips fixes.
the verifier contract — pasted by cclio into a `--bg` spawn prompt as `/x:verifier-brief <BYT-N|DOT-N> <pr url> <coder session id> <coordinator session id>`, one verifier per pr-lane coder. never auto-loaded.
Load when Dima says «walk me through» something, a ticket carries the walkthrough label, or teaching-by-showing beats a doc.
Load EVERY time a message for a human is about to be written — Dima says «write an email», «reply to this», «compose a message», «answer the recruiter», pastes a message asking for a response, or an outward-facing text (health update, announcement, readme prose) is about to ship. Pairs with matt's writing-for-agents: that one writes for machines, this one for people.
Load whenever a YouTube link is pasted, or Dima asks to transcribe / summarise / ingest a video, or asks what a video already on the shelf said.
Plugin manifests1
{
"$schema": "https://anthropic.com/claude-code/plugin.schema.json",
"name": "x",
"version": "0.11.99",
"description": "Dima's own skills — code core values + language guides, CST session handoff, commit workflow, RL task authoring, prompt review, agentic loops, pm ticket quick-ops, cross-repo conventions",
"author": {
"name": "Dima Vakatsiienko",
"email": "[email protected]"
},
"homepage": "https://github.com/dvakatsiienko/frame",
"repository": "https://github.com/dvakatsiienko/frame",
"license": "MIT"
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[x on Agent Plugins Marketplace](https://pluginsmp.com/plugins/x)