agent-skills

Autonomous multi-step deep research engine implementing an iterative Think → Search → Extract → Synthesize → Stop loop. The LLM drives every decision: what to search, what's relevant, what's missing, and when to stop. Produces a cited, magazine-quality report with inline citations, category- specific formatting, and research stats. Trigger when the user asks for "deep research", "research report on", "comprehensive analysis of", "look into X in depth", "write a report on X", or any question needing multi-source synthesis beyond a single search. For entity vetting/dossiers use entity-research; for news digests use news-monitoring; for source-grounded Q&A use notebooklm-mode.

Track upstream drift and sync adapted skill libraries

Use when the user wants to organize, tidy, or restructure a messy directory (Downloads, Desktop, documents folder). LLM-powered file organizer that scans content, proposes a structure, and executes moves in chunks with a mandatory user confirmation gate.

Compile a task brief BEFORE starting substantial work — pin down the goal (what "done" looks like), the context (everything known that bears on the task), the constraints (guardrails, with the organisation's standing rules included automatically) and the tooling (which installed skill or tool will do the work) — confirm it with the user, then execute against it. Use whenever the user says "brief this task", "brief this first", "make sure you understand before you start", "scope this properly", "compile a brief", "what do you need from me", "improve my prompt", "help me get better output", or hands over a SUBSTANTIAL task — multi-step, multi-document, or producing a deliverable someone else will read — as a thin one-line request. Asks at most 2-3 clarifying questions, and only where the answer would change the output; otherwise it states its assumptions in the brief and proceeds. Not for quick one-step asks — just answer those directly.

Convert Claude Code plugins into self-contained Hermes plugins — discovery analysis then full conversion

Parse agent log files to identify error patterns, rate limit hits, timeout clusters, tool failures, and component-level error counts. Produces a structured anomaly report. Cron-compatible — silent if no issues, alert digest if anomalies found.

Generate short marketing/social VIDEO clips via fal.ai through a staged studio workflow — brainstorm the motion, draft cheaply, then produce the final at quality. Three modes: text-to-video (a clip from a prompt, no source image); animate (bring a still image to life — a product, people, b-roll, a space); and camera-move (a gentle push-in, pan, tilt or shallow orbit over a still, including an interior "3D view"). Use when the user wants to "make a video/clip", "generate a video from text", "animate this image/photo", "create a marketing reel / social video / teaser", "do a 3D / parallax move", "pan/zoom/orbit a shot", or when they mention fal.ai, Kling, Veo or Seedance for video. Do not use for still images (use image-studio), Canva-style template designs, slide decks, data charts/dashboards, or flowcharts/diagrams. Carries an honesty discipline for real subjects: never use text-to-video to depict a real, identifiable place, product or person. Do not override platform-native video tools where the host environ

Generate, edit and upscale AI images via fal.ai through a three-stage studio workflow — brainstorm a strong prompt with the user, prototype cheaply and iterate on feedback, then produce a finalised image. Use when the user wants image generation or editing through the local fal.ai helper workflow, including requests to "generate an image", "make an image/picture/illustration/graphic of…", "create an AI image", "edit/change this image", "make a variation", "upscale this", create a "production-ready image", "clean up / enhance a photo", "make this phone shot look professional", create imagery "for the deck/post/website/newsletter", or when they explicitly mention fal.ai or nano-banana. Do not use for Canva template designs, branded PowerPoint decks, data charts/dashboards, or flowcharts/diagrams — those are layout, data, or structure tasks, not generative imagery. Video generation is out of scope. Do not override platform-native image generation tools where the host environment requires them.

Search and download free real-world stock photos from the Pexels API. Use when the user wants a REAL photo — "find a photo of X", "search for pictures of Y", "stock image of Z", "I need a photo for this slide/article/presentation". Searches the Pexels library (millions of photos) and downloads images in the chosen size and orientation. Do NOT use for AI-generated art ("generate an image of", "create a picture of", "make me an illustration") — those go to the host's image generation tool or an AI image skill. Key distinction: "photo" = real stock photography = Pexels; "generate/create/make an image" = AI art = image generation tools. Also do not use for editing existing images, screenshots, diagrams, or data charts.

Blind side-by-side multi-model comparison. Send one prompt to 2-4 models simultaneously, present responses anonymously (Model A / B / C / D), let the user pick a winner, then reveal identities and show which model won. Supports custom evaluation criteria, synthesis of responses, and vote history logging. Trigger when the user says "compare models", "test these models", "which model is better for", "A/B test", "blind comparison", "model evaluation", or wants to see how different AI models handle the same prompt. Can also be used for prompt engineering — testing how different models interpret the same instructions.

ADR-style decision journal for agents and teams. Create numbered decision records, track superseding chains, schedule periodic reviews, and search past decisions to avoid re-litigating settled questions.

Convert between document formats: Markdown↔HTML, CSV↔JSON, YAML↔TOML, JSON↔YAML, CSV↔Markdown table, HTML→plain text, JSON→CSV (flattening nested objects), Excel→CSV, Markdown→PDF. Requires pandoc for PDF and openpyxl for Excel.

Bulk-fill ONE master template — a Word (.docx) letter/form or an Excel (.xlsx) form — from a data table, producing one filled file per row (mail-merge). Use whenever the user wants to "fill in this letter for each person", "mail merge", "generate letters for this list", "run this template over a spreadsheet", or hands over a template plus a list. The skill reads the master, proposes a TOKENISED version (varying parts become {{tokens}}), confirms the template and the token-to-column mapping with the user, then regenerates one output per data row — preserving the master's layout and styling exactly. Data is an .xlsx or .csv (one row per output); outputs are one named file per record. Never invents: a token with no data is written as a VISIBLE flag, never a silent blank. Runs fully local and generates files only — it does not send, post or sign. Not for extracting data OUT of documents, and not a substitute for a hand-crafted single letter.

Cron-driven cross-session digest. Aggregates session activity, cron job outputs, memory changes, and tool usage stats into a compact summary for delivery via messaging platforms. Surfaces outstanding tasks and cross-session context that's invisible on chat platforms.

Create, update, sanitize, route, and export structured business-trip itineraries from booking confirmations, emails, tickets, PDFs, screenshots, calendar invites, or notes. Use for day-by-day itinerary generation, privacy-safe sharing, route links, and Markdown/PDF/DOCX/Google Doc/ICS exports.

Deep background research on an entity — a company OR a person — into a cited dossier: identity & background, ownership & key management, adverse / negative media, public sanctions-list name-match signals, PEP indications from public research, and litigation / regulatory history. Use when the user says "research this company / person", "background check on X", "any negative press / adverse media on X", "who owns / who runs X", "is X sanctioned / sanctions check on X", "vet this vendor / counterparty / candidate / partner", or "entity search". It RESEARCHES and COMPILES with sources; it is NOT a compliance/AML/CDD determination and NOT a sanctions or PEP clearance. A sanctions-list or PEP signal is a SIGNAL to escalate to a human compliance function, never a "clear" or "block". Research a person only for a legitimate purpose and only from public information.

Targeted claim verification pipeline. Given a factual assertion, search multiple independent sources, cross-check for agreement or contradiction, rate confidence (verified / likely true / disputed / unverified / outdated), and produce a cited verification report.

Semantic search over a personal library using bge-m3 embeddings + sqlite-vec. Index books, documents, any text corpus; query by meaning. Includes EPUB→Markdown conversion and MCP server for auto-available search tools.

Technique-focused media analysis. Detects rhetorical tools (loaded language, cherry-picking, source selection bias, framing, omission, emotional appeals, false balance) in articles and produces a structured analysis report. Identifies specific techniques and bias-signal intensity without labeling political positions.

Recurring topic/news monitoring with web search, multi-language sources, digest formatting, and automated delivery via Hermes cron jobs. Covers search strategy, source selection, Chinese-language platforms, and digest templates.

Source-grounded research pipeline inspired by NotebookLM. Collect verbatim source extracts into a temporary workspace, then answer questions and create deliverables grounded in those sources. Two grounding modes: strict (vault-only) and augmented (vault + labeled [background] knowledge). Trigger when the user asks to research a topic from sources, wants NotebookLM-style grounded answers, asks to build a resource vault, or says "notebooklm mode".

Enrich and search People Data Labs (PDL) Person and Company data, writing results to .xlsx. Five operations: enrich named PEOPLE into a profile, employment history and LinkedIn URL; IDENTIFY an ambiguous person as several scored candidates; SEARCH people by criteria (company, title, location); enrich COMPANIES into firmographics (industry, size, employees, HQ, founded, LinkedIn); and SEARCH companies by criteria. Use when the user has a list of names or companies and wants their job, employer, work history, LinkedIn URLs or company profiles looked up, or wants to find people or companies matching criteria. Trigger on phrases like "enrich these contacts", "find their LinkedIn", "look up these people", "who works at X", "find directors at Y", "enrich these companies", or "get firmographics for". Do not scrape LinkedIn directly — this uses a licensed data aggregator (PDL) instead, for legal and reliability reasons.

Persistent citation database for multi-session research. Add URLs as they're cited, dedup variants, tag by topic, check link health, and export bibliographies in Markdown/BibTeX/CSV/JSON.

Find and summarize YouTube videos for topics where visual explanation, demos, tutorials, talks, walkthroughs, or screen recordings are useful. Can run standalone, or export transcript-backed video source notes into notebooklm-mode vaults for grounded research.

Generic playbook for extracting structured data from any website — hotel prices, flight fares, e-commerce listings, real estate, jobs, competitor product catalogues, anything where the goal is to turn one or more URLs into clean records on disk. Use this skill whenever the user mentions scraping, harvesting, extracting, or pulling data from a website; names a specific site or competitor they want data from; asks how to handle JavaScript-rendered pages, Cloudflare blocks, bot detection, Turnstile challenges, or Playwright; wants to monitor prices over time; needs to automate a copy-paste research task; or says things like "just get the data from X" or "I need a script that grabs N from site Y". Covers recon, picking the lightest extraction tool that works, surviving anti-bot defences, and writing clean JSONL output with a run manifest. Scope is scraping only — landing data in a database, scheduling, and downstream pipeline work are explicitly out of scope and live in other tools.