Prompts, agent workflows, memory and tooling for LLMs.
Find which skill or flow in this repo fits your current situation.
Diagnose AI agent failures with a reproducible capture and report, not a retry.
Point to matching Claude Academy courses when someone asks how to use Claude.
Find, retrieve and improve AI prompt templates from prompts.chat.
Operate a Dify server and build, test and publish its apps with difyctl.
Build ChatGPT widget apps with mcp-use; superseded by mcp-app-builder.
Explains when to hand code location, small edits or diff review to compressed agents.
Automates a web browser and returns AI-optimized page snapshots for agents.
Sends tasks to the OmniRoute agent-to-agent server and tests its JSON-RPC protocol.
Runs the adaptive-coordinator agent role from the ruflo multi-agent toolkit.
Designs and builds AI agents, covering their architecture, tools, subagents and planning.
Automates Ably tasks through the Rube MCP connector.
Investigate why a session went wrong and draft a bug report for maintainers.
Pursue a long goal in one session by coordinating subagents.
Trim a repo's agent setup into daily and library buckets using evidence.
Guides building AI agents with the Cline SDK, covering runtime, tools, plugins, and teams.
Automates websites for an AI agent: clicking, filling forms, screenshots, and scraping.
Mines projects and chats into a searchable memory palace for recalling past work.
Records work passed between bots in shared memory so the next bot can pick it up.
Answers BMad questions and recommends the next installed skill to use.
Runs interactive onboarding that takes a new member to one confirmed issue-based task.
Configures and runs DeepTutor from the command line, including knowledge bases and sessions.
Answers Cherry Studio feature questions and diagnoses runtime errors from the installed build.
Searches and installs agent skills from the ClawHub public registry.
Diagnoses an AI project and proposes existing skills and a written implementation plan.
Controls Herdr panes, tabs, and workspaces to run commands or drive other agents.
Approves pairings and edits the allowlist and DM policy for the Discord channel.
Automates websites for agents: navigating, filling forms, clicking, and screenshots.
Search and install agent skills from the prompts.chat registry.
Talk to the OpenClaw agent through its Discord-backed session.
Turns your own course-making history into a reusable personal skill.
Splits a large goal into bounded units, runs them in parallel, and verifies the result.
Sets up a CopilotKit agent that answers in Slack or Microsoft Teams end to end.
Distills a person or topic into a runnable thinking-framework skill after research.
Converts books and documents into structured agent skills with frameworks and principles.
Drives a real browser in the sandbox to open pages, fill forms, and extract data.
Run independent tasks with separate agents when they share no state.
Pass the current conversation to a fresh background agent that continues the work.
Answers questions about OpenAI APIs, Codex settings and model choice from docs.
Routes file reads, searches, and data-heavy commands through context-mode tools.
Chooses and opens browser identities for web tasks, and when a plain download is enough.
Route and complete Composio work across Composio For You and Composio Platform.
Automates websites through the agent-browser CLI: navigation, forms, screenshots, and data.
Wires agentmemory into host coding agents and checks their config paths.
Audits an agent system prompt against a rubric of contract, fallbacks, and anti-patterns.
Builds, grades, and schedules Claude Managed Agents in your own account.
Saves, recalls, and organizes notes across sessions when asked.
Explains how to write a clear prompt when delegating work to a subagent or fork.
