493 skills · sorted by relevance
Turn an unclear feature idea into a structured specification with goals, boundaries, and testable success criteria. Split broader requests into independent modules when needed, then move through planning, task breakdown, and implementation with human review.
Prepare media assets required by an existing OD Next plan. Reuse verified capabilities and suitable assets, acquire files, measure dimensions, and retrieve background job results while preserving required content, licensing constraints, and quality conditions.
Increase image and video resolution using AI super-resolution models hosted on fal.ai. This catalog entry helps agents discover the capability; running the full workflow requires installing the upstream bundle with its original assets, scripts, and references.
Automatically repair source code against a failing test while preserving the test itself. Set a spending limit and attempt count, then receive verification results and usage costs based on rerunning the test command.
Compare ruflo and HAL systems for running the GAIA benchmark, an AI evaluation. Examine differences in code execution, browsing, file handling, memory, and model selection to prioritize improvements by expected impact and effort.
Review a reusable Agent Skill package before implementation. Validate its identity fields, separate instructions from external capabilities and execution timing, and receive a validation report with a minimal set of building blocks and reasons for each choice.
Explain how dot works and check why a task is blocked, including computer access, login, approvals, and Slack errors. It provides supported next steps and answers questions about changing dot’s avatar and the user’s message color.
Create a new native Muse plugin package with its manifest and requested capability files, then validate it using the installed muse CLI. Receive a file inventory, validation results, and an installation proposal for later use.
Discover what a paired phone or wearable currently supports, then run the requested device, app, camera, media, or smart-home action. Check each device's published capabilities and permissions, and report the actual outcome or any execution errors.
Add interactive features to published Artifact pages using supported runtime capabilities, features granted by the viewer environment. Build saved checklists, shared data views, connector integrations, Claude-powered interactions, and file uploads or downloads with permission-aware behavior.
Plan image and graphic design workflows by matching tasks to available tools, from AI generation and photo editing to logos, interface mockups, diagrams, and publishing. Compare requirements, costs, and offline options before combining providers.
Plan game development workflows by matching engines, art and audio tools, automated playtests, packaging, publishing, and analytics to available tools and credentials. Compare options by cost, quality requirements, and offline availability.
Find agent skills for a specific task, compare candidates using their capabilities, sources, and installation history, and install a selected skill into Cherry after the user has expressed an intent to install it.
Reviews Cherry Studio code and documentation against product decisions, architecture rules, implementation checks, and conventions. It selects a review approach based on scope and available agents, then reports findings with guidance for fixes.
Inspect an agent’s current model, processing limits, workspace, tools, and message routing details. Diagnose unavailable capabilities, adjust permitted runtime settings, or keep temporary session notes, with advance notice before changes that affect the model or limits.
Compare AAS skills against a project's capabilities, then record the chosen combination as a validated stack manifest. Inspect the result, identify catalog gaps, and optionally export evidence documenting the selection process.
Connect an AI agent to over 100 cryptocurrency exchanges and prediction markets through a local MCP server. Fetch prices, order books and account balances, or place and cancel orders when trading is explicitly enabled.
Add a new exchange integration to CCXT using TypeScript and established exchange templates. Follow guidance for unified methods, data conversion, authentication, sandbox settings, recorded API test data, documentation, and checks across generated language implementations.
Examine container isolation in a CTF sandbox, a controlled security challenge environment. The skill checks kernel and runtime conditions, verifies whether an exploit crosses into host context, and preserves a minimal reproducible chain with evidence.
Trace SSH keys, tokens, and other credentials in Linux CTF environments. Connect each recovered artifact to the service that accepts it, then document the resulting privileges or movement to another host.
Evaluate whether a fine-tuned model is ready for release by checking training data, general capability loss, comparisons with the base model, and staged production rollout. Produces an evidence-based promotion or rejection report with one prioritized remedy when rejected.
Choose a model training method and configuration from paired response preferences or individual positive and negative ratings. Covers DPO, ORPO, KTO, and SimPO, including pair construction, repeated training rounds, and troubleshooting learning settings.
Finds a local MCP server, a connector that adds tools to an agent, when existing tools or simple code cannot handle a task. It selects packages requiring no environment variables and starts them after approval.
Convert Simplicité text logs into structured JSON while keeping only the fields needed for troubleshooting. Preserve multiline errors and stack traces, and reduce the amount of log content passed to an AI agent for analysis.