Searches the material already in reps' hands for the pieces and the claims that a change puts in doubt, and lists what has to be revisited.
Use when the long-term vision is big but the entry point is fuzzy — find the credible first wedge that expands.
Search markdown files, Nextflow scripts, and config files for TODO/FIXME comments, categorize by priority, and provide actionable recommendations.
Search markdown files, Nextflow scripts, and config files for TODO/FIXME comments, categorize by priority, and provide actionable recommendations.
Use when the user needs to find a tool, library, skill, MCP server, OSS project, or paid SaaS that solves a specific task.
Analyze the codebase to discover and extract the ubiquitous language. Identify domain terms, their definitions, and how consistently they are used.
Audit and clean the Remotion GitHub masterplan hierarchy. Use when finding open issues that lack a masterplan ancestor, identifying orphaned issues that need a parent, or…
Use when user wants to find, look up, or resolve a Telegram user by username, phone number, or ID, or get someone's Telegram profile info.
Pipeline de debug em 3 fases com 5 agentes para caçar bugs num componente e em todos os seus componentes relacionados (hooks, services, edge functions, tabelas).
Identify potential B2B clients matching service definition and ideal client profiles
Open Finder at relevant location. Use when user says open folder, show in Finder, or покажи файл
Für Findet: Erstprüfung, Rollenklärung und Mandatsziel: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt.
Manage GitHub project findings — create issues, list open items, or pick up work. Usage: `/finding` (create from the current conversation), `/finding list` (show open), `/finding…
Use when APIs fail repeatedly with version-related errors (method not found, wrong arguments, unknown flag) or when about to use library APIs with uncertain knowledge - g — from…
Detect profitable arbitrage opportunities across CEX, DEX, and cross-chain markets in real-time. Use when scanning for price spreads, finding arbitrage paths, comparing exchange…
Use when the user wants to work with a single radiology finding interactively — checking whether it exists in defs/, adding a synonym to an existing model, or drafting a new…
How to find, evaluate, and recommend the best ingestion method for changelog URLs — covers feed discovery, provider detection, GitHub API, markdown sources, and scraping fallback
Finds similar implementations, usage examples, and existing patterns in the codebase - shows concrete code examples for modeling new work.
Use when reviewing code changes, auditing new features, cleaning up PRs, or user says "find dead code", "find unused code", "check for unnecessary additions", "what can I remove".
Use when auditing a codebase for semantic duplication - functions that do the same thing but have different names or implementations.
Finds and ranks expensive Snowflake queries by cost, time, or data scanned. Use when: (1) User asks to find slow, expensive, or problematic queries (2) Task mentions "query…
File-location search strategy for the file-finder agent — glob by naming convention, content search, import tracing, directory exploration, and manifest checks, scoped to the…
Help users build relationships with mentors and sponsors for career growth. Use when someone is looking for career guidance, wants to find a mentor, needs an advocate at work, is…
Finds the most informative session recording linked to an error tracking issue. Use when a user has an error tracking issue ID and wants to watch a replay showing what th — from…
Guides a user from "I want to watch recordings but don't know which ones" to a short, high-signal list of sessions worth watching.
Single source of truth for per-finding review-triage dispositions and response selection for actionable findings, which the review and audit skills apply by reference.
Use when optimizing existing pages for SEO, hunting low-hanging keyword opportunities, analyzing pages that get impressions in Google Search Console but few clicks, or planning a…
Use when starting ambiguous or high-stakes work, or when the user says "blind spot pass", "interview me", "what am I missing", or asks for an implementation plan.
Findity integration. Manage data, records, and automate workflows. Use when the user wants to interact with Findity data.
Track Apple devices and AirTags via FindMy.app on macOS using AppleScript and screen capture.
Track a shared contact's location via Apple Find My with street-level accuracy. Returns address, city, and context (home/work/out) by reading map landmarks.
Findymail integration. Manage Persons, Organizations, Leads, Deals, Pipelines, Activities and more. Use when the user wants to interact with Findymail data.
Automate Findymail tasks via Rube MCP (Composio). Always search tools first for current schemas. — from general/automation
Appeal a parking ticket, penalty charge, or administrative fine with the grounds that actually get appeals granted — not indignation.
Use when a task needs the judgment of a fine artist — pricing and raising prices on a body of work, evaluating a gallery representation or consignment offer, structuring a studio…
에어코리아 기반 미세먼지/초미세먼지를 지역명 또는 위치 힌트로 조회한다. 기본 경로는 k-skill-proxy의 report endpoint다.
Fine(맥 GUI)을 설치·점검·설정할 때 사용한다. 설정창이 없으므로 `fine` CLI로 상태를 읽고 빠진 것을 채운다. 하네스 추가, 모델·외형·배경 설정, 재시작, 문제 진단.
Determines whether to fine-tune vs. prompt-engineer, and if fine-tuning is justified, produces dataset requirements, training data format, pre/post eval plan, and cost-benefit…
Plan supervised fine-tuning, LoRA, QLoRA, or full-parameter training with reproducible evaluation gates.
Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support — from axolotl-ai-cloud/diff-transformer
LLM fine-tuning with LoRA, QLoRA, DPO alignment, and synthetic data generation. Efficient training, preference learning, data creation.
Use when fine-tuning LLMs, training custom models, or adapting foundation models for specific tasks. Invoke for configuring LoRA/QLoRA adapters, preparing JSONL training — from…
Use when fine-tuning LLMs, training custom models, or optimizing model performance for specific tasks.
Use when fine-tuning LLMs, training custom models, or optimizing model performance for specific tasks.
Use when fine-tuning LLMs, training custom models, or adapting foundation models for specific tasks. Invoke for configuring LoRA/QLoRA adapters, preparing JSONL training — from…
Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support — from Orchestra Research
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1 — from…
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization — from Orchestra Research
Decide between prompting, retrieval, and fine-tuning with eval-first discipline and honest data requirements. Use when someone proposes fine-tuning or a prompt has hit its ceiling.
Use when fine-tuning LLMs with TRL — SFT, DPO, PPO, GRPO, reward modeling, RLHF. Triggers: SFT, DPO, GRPO, fine-tune, RLHF, reward model, TRL.
Finerio Connect integration. Manage data, records, and automate workflows. Use when the user wants to interact with Finerio Connect data.
Automate Finerworks tasks via Rube MCP (Composio). Always search tools first for current schemas. — from general/automation
Turn raw examples into a training-ready fine-tuning dataset — normalize to the trainer's chat/instruction format, deduplicate (including near-duplicates), strip PII, balance,…
Use when preparing to fine-tune an LLM for multi-turn conversations, before generating any training data.
Use when generating synthetic training data for multi-turn conversation fine-tuning. Triggers - have design artifacts ready, need to generate conversations, ready to assess…
Teach a character/persona to a Qwen3 base model with LoRA, evaluate before/after, export to GGUF, and load it into a naia container via the model-swap API.
模型微调服务 | Model Fine-tuning Service. LLM LoRA/QLoRA 微调 | LLM LoRA/QLoRA fine-tuning. 7B/13B 模型微调 | 7B/13B model fine-tuning.
Use when training a fine-tuned model and evaluating improvement over base model. Triggers - have filtered training data, ready to submit training job, need to convert to GGUF.
Model fine-tuning with PyTorch and HuggingFace Trainer. Covers dataset preparation, tokenization, training loops, TrainingArguments, SFTTrainer for instruction tuning, evaluation,…
Decide whether to fine-tune at all, and route to the right method (SFT, DPO/ORPO/KTO, GRPO/RLVR, continued pretraining) and base model.