anton-abyzov

Find live X threads where a product genuinely fits, and draft truthful, humanized replies for the owner to review and post by hand. Human pace only: max 5 replies a day, never two within 20 minutes, no automated posting. Use when the user asks to 'find…

Manual sweep of a SpecWithCode project after an update, restart or outage: list every thread, continue only work the restart cut off and the app did not continue, never touch work a person stopped, hand off one thread at a time only through existing handoff…

Read-only exposure check and fix plan for php2shell, the October 2026 PHP emergency patch (CVE-2026-109618, CVE-2026-109619, unauthenticated RCE in every web SAPI). Runs a portable POSIX script that finds installed PHP binaries and versions, judges each…

Installed copy of anton-abyzov/greet-anton-test hello-skill at 1.0.0, used by the live update-pipeline E2E.

Get public web data the cheapest legal way, with Apify as the paid fetch layer: choose official API, plain fetch or an Apify Store actor, cap every run's spend, and save evidence. Activate when the user wants to scrape, crawl, extract, monitor or enrich data:…

Social media content creation, posting and engagement on Instagram, LinkedIn, X/Twitter, Threads, YouTube, TikTok, Reddit, dev.to, Facebook, Discord and Telegram: AI images (Nano Banana Pro, 3 options per image), AI video (Veo 3.1), carousels, per-platform…

iOS/macOS app deployment and App Store Connect automation using the `asc` CLI — a lightweight alternative to fastlane. Activates when deploying iOS apps, distributing TestFlight builds, submitting to App Store, managing certificates/profiles, uploading…

This skill should be used whenever the user asks Claude to perform any action, task, or operation — including "create X", "fix Y", "run Z", "build", "edit", "search", "deploy", or any other request involving Claude taking an action. Make sure to use this…

Review a proposed sharing or publishing workflow for a private AI skill using supplied repository visibility and authorization details. Identify intended recipients and missing access evidence before suggesting a sharing path.

Review supplied software release receipts and state exactly which source, test, deployment, installation and runtime claims they support. Use for release readiness or completion claims.

Compare supplied AI skill inventories from two machines or agents by source identity, version, content hash and installation scope. Use to find missing or changed skills without mistaking installation for execution.

Generate human-readable release notes grouped by type from a range of git commits.

Generate human-readable release notes grouped by type from a range of git commits.

Triage third-party dependencies before adoption — maintenance health, security advisories, supply-chain risk, and a go/no-go scoring rubric.

Audit a project's dependency licenses for compatibility and obligations — copyleft detection, license allowlists, and SPDX-based reporting.

Generate and maintain a Keep a Changelog style CHANGELOG.md from Conventional Commits, grouped by type with correct semver headings.

Run a SpecWeave increment unattended until every task is done, then verify and close - the Unattended section of sw-do. Use for "auto mode", "run until done".

Work a SpecWeave increment task by task - claim, implement inside its Files, test, commit, done with evidence - alone, unattended or in lanes, then close. Use for "implement", "continue", "auto".

Close a SpecWeave increment - every task done or skipped, specweave verify green, a fresh-context review, then specweave complete. Use for "close it", "we are done", "finish up".

Hand work to another tool, account or machine and pick it up there, in two words. Use when the user says "hand off", "hand off all", "out of tokens", "switching accounts", "pick up" or "continue".

Plan work as a SpecWeave increment - one spec.md with Problem, Scope, numbered ACs, Approach and Tasks - before writing code. Use for "plan a feature", "new increment", "let's build X".

Use Jev or optional OpenAI Decisions for bounded routing, shell-command risk, text screening and failure triage. Use for "jev", "system one", or "classify this".

Keep a project's goal, decisions, work items, artifacts and briefs portable across AI tools and accounts. Use for project coordination, "project brief", "what is this project about".

Adversarial review of an increment before it ships, by a fresh context or another model family; every finding cites path:line and is re-verified. Use for "review", "qa", "critique this".

Run a SpecWeave increment unattended until every task is done, then verify and close - the Unattended section of sw-do. Use for "auto mode", "run until done".

Work a SpecWeave increment task by task - claim, implement inside its Files, test, commit, done with evidence - alone, unattended or in lanes, then close. Use for "implement", "next task".

Close a SpecWeave increment - every task done or skipped, specweave verify green, a fresh-context review, then specweave complete. Use for "close it", "we are done", "finish up".

Interview the user relentlessly, in rounds, until every decision in a plan is settled; facts are your job, decisions are theirs. Use for "grill me", "interview me", "stress-test this plan".

Hand work to another tool, account or machine and pick it up there, in two words. Use when the user says "hand off", "hand off all", "out of tokens", "switching accounts" or "pick up".

Plan work as a SpecWeave increment - one spec.md with Problem, Scope, numbered ACs, Approach and Tasks - before writing code. Use for "plan a feature", "new increment", "let's build X".

Adversarial review of an increment before it ships, by a fresh context or another model family; every finding cites path:line and is re-verified. Use for "review", "qa", "critique this".

Adversarially review finished work that another model family built - hunt for bugs, risky assumptions, missed requirements and security problems, rank findings by severity with path:line evidence, and never rubber-stamp. Never edits, commits or pushes. Use…

Enforce SpecWithCode's PR contribution policy by closing ineligible submissions and triggering Macroscope review for eligible work within authorized scope. Supports explicit dry runs. Use for contribution moderation, not installation diagnostics or a full…

Test SpecWithCode's web and desktop UI through its built-in Browser panel against isolated development state. Use for browser verification, browser pairing recovery, and test fixtures. Use test-studio-mobile for native mobile verification.

Test SpecWithCode's native iOS and Android app through its Device panel and returned AgentDevice command. Use for mobile verification, native-client builds, Metro launch, and mobile pairing against isolated development state.

Write a git commit message for a diff in this team's format. Use when asked to write, suggest or fix a commit message.

Work a SpecWeave increment task by task - claim, implement inside its Files, test, commit, done with evidence - alone, unattended or in lanes, then close. Use for "implement", "continue", "auto".

Hand work to another tool, account or machine and pick it up there, in two words. Use when the user says "hand off", "hand off all", "out of tokens", "switching accounts", "pick up" or "continue".

Plan work as a SpecWeave increment - one spec.md with Problem, Scope, numbered ACs, Approach and Tasks - before writing code. Use for "plan a feature", "new increment", "let's build X".

AnyModel development skill — the universal AI coding proxy. Use this skill whenever working on AnyModel source code (proxy.mjs, cli.mjs, providers/*, site/*, test/*), adding providers or presets, debugging proxy issues, deploying to npm, or modifying the…

AnyModel development skill — the universal AI coding proxy. Use this skill whenever working on AnyModel source code (proxy.mjs, cli.mjs, providers/*, site/*, test/*), adding providers or presets, debugging proxy issues, deploying to npm, or modifying the…

This skill enforces a strict pre-response greeting protocol where the literal string 'HI Anton Abyzov' must appear at the very start of every reply to every user prompt — no exceptions. Activate whenever a persistent session prefix greeting is required, or…

This skill enforces a strict pre-response greeting protocol where the literal string 'HI Mr. Anton Abyzov' must appear at the very start of every reply to every user prompt — no exceptions. Activate whenever a persistent session prefix greeting is required,…

Write a portable, secret-scrubbed work handoff doc so you can continue this work in any AI tool. Use when saying "handoff", "running out of tokens", "switch to Codex/OpenCode/Gemini/Cursor", "continue elsewhere", or "continue on another machine".

End-to-end corporate and personal tax preparation: data gathering from Xero/bank statements/Gmail/Obsidian/Google Drive, P&L generation, IRS compliance analysis, tax calculation, document staging, and payment guidance. Use this skill when the user mentions…

One skill for creating, scheduling, posting and replying on social media, with explicit per-post approval. Covers routing (X and TikTok via Blotato, everything else via self-hosted Postiz), account and integration IDs, YouTube channel routing, Postiz/Blotato…

Short-form video and ad production plus the Postiz internal-API mechanics behind it, for Anton's self-hosted Postiz (postiz.easychamp.com). Covers cookie-auth Postiz calls (showorg and User-Agent headers, upload, per-platform settings, in-place update and…