Anton Dziatkovskii

Alias for the retro skill: identical behavior, shorter to type. On any trigger, run the retro skill, which holds the single copy of the logic. Triggers: "/rr", "rr", "retro".

中文简介正在准备中。当前显示原文说明。

- End-to-end passive reconnaissance for a domain, website or IP — builds an asset inventory covering registration, DNS, subdomains, infrastructure, tech stack, history and ownership without sending a single packet to the target. Use when asked to research or profile a domain or website, map what an organisation runs, or investigate a suspicious site without alerting its operator

中文简介正在准备中。当前显示原文说明。

Audit and cut the cost of a machine's scheduled routines: measure tokens per run, pick a lever (cheaper model, split, lower frequency, shorter prefix, retire) and prove the saving with a before-and-after measurement. Use when background routines burn more than interactive work. Triggers: /routine-opt, /ropt, routine audit, cheaper routines.

中文简介正在准备中。当前显示原文说明。

Finds high-value SEO and GEO keywords using web search, AI analysis, and optionally paid tools like Ahrefs or Semrush. Produces a validated keywords.csv file with a fixed schema for downstream pipeline consumption.

中文简介正在准备中。当前显示原文说明。

Report this machine's file-synchronization state with the whole fleet: peers connected, per-share state and files still needed, folder errors, and how many sync-conflict files piled up, which is the silent sign two machines are fighting over a file. Read-only and zero tokens. Triggers: "/sync-check", "is sync alive", "sync status".

中文简介正在准备中。当前显示原文说明。

Work leads in Telegram: find prospects by topic, keep only those who self-mentioned it, resolve their handle including through shared groups, pitch personalized and grounded in real templates, run a two-step scheduling close, triage follow-ups, and capture everyone in the CRM. Triggers: "find leads for <topic>", "pitch this prospect", "run outreach".

中文简介正在准备中。当前显示原文说明。

Учебные аудиоподкасты для Антона в NotebookLM: выбирает, чему учить (труба собеседований, дыры сверки), пишет пакет и промпт ведущим, генерирует аудио RU+EN через CLI notebooklm, кладёт в волт, ведёт учебный план и YouTube-очередь. Триггеры: /teach-podcast, /teach, сделай подкаст, обучи меня, подкаст в машину, что мне послушать. НЕ /notebooklm (общий слой артефактов).

中文简介正在准备中。当前显示原文说明。

Delegate work over a shared task chat: a task is assigned by a short code phrase, and the assignee's agent expands it into a full seed and starts immediately, with state tracking so no delegated task dies silently. Triggers: "/task", "task <ID>", "my delegated tasks", "delegate <task> to <person>".

中文简介正在准备中。当前显示原文说明。

Stop a Google Takeout export from expiring undownloaded: detect the ready-to-download mails with a deterministic zero-token scan, report the live links with days remaining, then drive the download and vault import. Triggers: "/takeout-pull", "is my takeout ready", "pull my takeout".

中文简介正在准备中。当前显示原文说明。

Review content quality before anything is shown or sent (vault notes, outgoing drafts, dedup merges, service files) and return an explicit pass, fail or manual-review verdict. The counterpart of a post-build test gate: that one asks whether the build works, this one whether the writing is fit to show. Triggers: "/taste-check <path|text>", "/taste", "check this note".

中文简介正在准备中。当前显示原文说明。

Query the user's local and synced-device data via the screenpipe REST API at localhost:3030 — recordings, audio, UI, meetings, connected services, and memory. Use for screen activity, other-device or cross-device history, productivity, media export, connections, durable memory, or discovering and automating saved workflows.

中文简介正在准备中。当前显示原文说明。

Run a retrospective for an abandoned agent session (untouched for over 24 hours, never closed with a retro): rebuild context cold from the transcript, list what was built, sort Keep/Drop/Try and file the durable parts. Use from a nightly routine or on demand. Triggers: /retro-lost, retro for an abandoned session.

中文简介正在准备中。当前显示原文说明。

Write a public post in the owner's own voice with a role model's style and idea palette layered on top, reading a style-palette file so the same engine works for any thinker the owner has profiled. Draft-first. Triggers: "/speak-as <name>", "write in the style of <X>".

中文简介正在准备中。当前显示原文说明。

Operating manual for Telegram over its MCP connector and roughly 115 tools: how to find a chat by title, the tool inventory grouped by job, resolving chat ids, auto-transcribing voice notes, and the connector gotchas (pinned-first ordering, big-page crashes, reload after reconnect, untrusted fields, always passing an explicit account). Load before any non-trivial Telegram work.

中文简介正在准备中。当前显示原文说明。

Сделать сессии Claude Code на ЭТОМ компьютере видимыми под любым аккаунтом, как локальные треды Codex. Триггеры: /sync-sessions, «синхронизируй сессии», «запейни сессии», «после смены аккаунта пропали чаты», «не вижу старые сессии».

中文简介正在准备中。当前显示原文说明。

Close a work session properly: reconcile it against the wider collaboration, inventory what was actually built from git log and recently-touched files, summarize the arc, classify each artifact as keep, one-off or promote-to-permanent, route the durable ones to memory, canon or a new skill, save the retro as a vault note, and hand back an enriched /compact block so working memory shrinks without losing the thread. Triggers: "/retro", "/rr", "wrap up the session".

中文简介正在准备中。当前显示原文说明。

Audits a website codebase and makes code changes so AI engines (ChatGPT, Claude, Perplexity, Google AI Overviews) can better discover, parse, quote, and cite the site. Covers structured data, content structure, technical signals, and freshness.

中文简介正在准备中。当前显示原文说明。

Protocol for two LLM agents (Claude, Codex or any other) to agree and work from one shared folder: message format, channels, roles, evidence rules, wake-up alarm and anti-patterns. Triggers: /llm-pact, agree with codex, connect a second agent, pair rules, agent consensus, two-agent protocol.

中文简介正在准备中。当前显示原文说明。

Short alias for /llms: settle the current question through a targeted, bounded consensus of other LLMs. Use when you want outside models to agree or disagree with a plan before acting. All logic lives in the llms skill. Triggers: /LLLL, llll.

中文简介正在准备中。当前显示原文说明。

Sweep every external pointer to you (profiles, READMEs, bios, link pages) after a rename of a repo, org, domain, handle or account, and fix the dead ones. Use right after any public rename and once a quarter as hygiene. Triggers: /link-rot, /lr, check links, broken profile links, after the rename, link rot.

中文简介正在准备中。当前显示原文说明。

Pick up and continue a serialized build-in-public book: read the current state (which days are written in which languages, what is uncommitted, where the gaps are), then keep writing through the established pipeline. Triggers: "/journey", "the book", "continue the journey".

中文简介正在准备中。当前显示原文说明。

- Verify whether an image or video is authentic, original and correctly captioned — provenance checks, error level analysis, noise and JPEG compression analysis, clone and copy-move detection, lighting and shadow consistency, C2PA… Use when fact-checking a photo or video, checking for a deepfake or AI-generated image, spotting manipulation, or testing whether footage is recycled or miscaptioned

中文简介正在准备中。当前显示原文说明。

- Investigator OPSEC — threat-model who might notice you, control your attribution surface across IP, ASN, browser and TLS fingerprint, timing and logged-in accounts, separate research identity from… Use when setting up a research account, avoiding tipping off a subject, worrying about LinkedIn profile-view leakage, needing a burner phone or email, or hardening a research VM or browser profile

中文简介正在准备中。当前显示原文说明。

- Start-here router and tradecraft baseline for any investigation into a person, company, domain, image or selector.

中文简介正在准备中。当前显示原文说明。