Loading...
Loading...
Found 452 Skills
Before searching a codebase, forces you to zero in on the target: what exactly are you looking for, what would it look like, where would it live, what else might it be called. Activates on "find", "where is", "search for", or when exploration begins. Prevents grep-and-pray.
Database operations including querying, schema exploration, and data analysis. Activates for tasks involving PostgreSQL, MySQL, MariaDB, SQLite, MongoDB, Redis, Elasticsearch, or ClickHouse databases.
[Fix & Debug] Investigate and explain how existing features or logic work. READ-ONLY exploration with no code changes.
Generate images using ModelScope Z-Image models (Z-Image-Turbo, Z-Image, Z-Image-Edit). Use when user asks to generate images, create artwork, or requests image generation functionality. Supports async generation with polling and optional LoRA configurations. IMPORTANT - Model Selection Rule: If the user explicitly mentions "Z-Image-Turbo" in their prompt, use "Tongyi-MAI/Z-Image-Turbo"; if they explicitly mention "Z-Image" (without Turbo), use "Tongyi-MAI/Z-Image"; otherwise, use the default "Tongyi-MAI/Z-Image-Turbo".
Fine-tune LLMs using the Tinker API. Covers supervised fine-tuning, reinforcement learning, LoRA training, vision-language models, and both high-level Cookbook patterns and low-level API usage.
Train custom AI models (LoRA) on fal.ai — personalize image generation for specific people, styles, objects, or video generation. Use when the user requests "Train model", "Train LoRA", "Fine-tune", "Custom model", "Train on my images", "Portrait training".
Systematic exploratory QA testing of web applications — find bugs, capture evidence, and generate structured reports
Explore-lane experimental execution skill for deep learning research repositories. Use when the researcher explicitly authorizes exploratory runs such as small-subset validation, short-cycle guess-and-check, batch sweeps, idle-GPU search, or quick transfer-learning trials, with results summarized in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted baseline execution, conservative training verification, default routing, or implicit experimentation.
Apply organizational ambidexterity theory to balance exploration and exploitation activities. Use this skill when the user needs to diagnose whether an organization is over-exploiting or over-exploring, design structures that support both innovation and efficiency, or evaluate the tension between short-term performance and long-term renewal.
Conduct targeted code exploration on a repository, and document the process of "Asking Questions → Reading Code → Reaching Conclusions" as searchable evidence for direct reuse when similar questions arise next time. There are three types: question (investigate code around a specific problem and provide conclusions), module-overview (organize the structure, boundaries, entry points, and dependencies of a module), spike (conduct lightweight technical exploration of multiple possible directions without making final decisions). Trigger scenarios: When users say "Let's explore first", "How is X implemented in this repository", "Quickly get familiar with this module", "Archive the exploration results". For the distinction from learning / tricks / decisions, refer to the root skill `easysdd`.
Adaptive exploration pipeline that integrates /brainstorm, /think, and /red-team with intelligent pivoting. Unlike /deepthink (which takes a fixed idea and iterates), /prospect starts with divergent brainstorming, picks the most promising vein, runs deep analysis, and — crucially — can PIVOT back to divergent thinking when: the idea dies under red-team, an adjacent opportunity surfaces during analysis, or the research reveals the real opportunity is elsewhere. Produces a prospecting report: the landscape explored, veins assayed, pivots taken, and the final stake with conviction. Use when the user says "prospect", "explore this space", "find opportunities", "what should I build", "explore and analyze", or has a domain/trend they want to both explore AND evaluate.
Documents the results of a time-boxed technical or design exploration (spike). Use after completing a spike to capture learnings, findings, and recommendations for the team.