AI & ML

drjmf Bundle

Clones a specific node (logic and visual representation) from one RagFlow agent to another using unique IDs to ensure UI visibility.

drjmf Bundle

Creates or updates a categorization node in a RagFlow agent using definitions (categories, examples, and descriptions) from an external JSON file. Ensures both logic and UI list items are synchronized.

fhgomes Bundle

Test engineering specialist. Use ALWAYS when conversation involves: writing, adding or fixing tests, "make this testable", coverage, flaky tests, red CI, Playwright / JUnit / Mockito / Testcontainers / MockMvc / vitest / Testing Library / Flutter widget tests, TDD, "is this tested?", "why is CI red", a pre-commit or agent hook that runs tests, mocks and fakes, unit vs integration, test data and teardown, "the AI says done but nothing is tested", or any variation of "how do I test X". Also trigger when user mentions: "spec", "suite", "assertion", "flaky", "green build", "regression test", "characterization test", "smoke test", "E2E", "mock", "stub", "fixture", "teardown", "golden test", "mutation testing", "quality gate", "skip-tests", "@Disabled", "test.skip", "test.only", "single-test command", "known-red baseline". If in doubt whether the task needs a test, trigger.

ag2ai Skill

Pause an AG2 beta `Agent` mid-run to collect human input via `context.input()`, or gate a tool call with `approval_required()` middleware. Use when the user wants the agent to ask for confirmation, request missing info (passwords, API keys, data), or have a human approve sensitive / irreversible / expensive tool calls (sending emails, deleting records, payments).

ag2ai Skill

Test AG2 beta agents and tools without hitting a real LLM provider. Pass `TestConfig(...)` from `autogen.beta.testing` as the agent's config (or per-`ask`) to mock LLM responses, inject `ToolCallEvent`s to simulate tool execution, and assert success / error paths. Use when the user is writing pytest tests for an Agent or Tool.

ag2ai Skill

Map of AG2 beta capabilities and which sibling skill to reach for. Load first when the user mentions building with AG2 beta (autogen.beta) but the specific feature isn't yet clear — agents, tools, model config, delegation, memory, observers, structured output, HITL, AG-UI, telemetry, or testing.

ag2ai Skill

Add OpenTelemetry traces to an AG2 beta `Agent` via `TelemetryMiddleware` (`autogen.beta.middleware.builtin`). Emits spans for the full turn, each LLM call, each tool execution, and each human-input request, following the OpenTelemetry GenAI semantic conventions. Compatible with any OTLP backend — Jaeger, Grafana Tempo, Datadog, Honeycomb, Langfuse. Use when the user wants production-grade traces, latency analysis, token-usage attribution, or to ship telemetry into an existing observability stack.

ag2ai Bundle

Build a minimal AG2 beta `Agent` end to end — pick a model provider, set a prompt, call `agent.ask()`, then continue the conversation with `reply.ask()` (multi-turn). Use when the user is starting a new AG2 beta project, has no working `Agent` yet, or needs the multi-turn chaining pattern. Covers `OpenAIConfig`, `AnthropicConfig`, `GeminiConfig`, `OllamaConfig` etc., and env-var fallback for API keys.

ag2ai Skill

Give an AG2 beta `Agent` the ability to run shell commands. Covers `LocalShellTool` (client-side `subprocess`, works with any provider) and the provider-native `ShellTool` (Anthropic / OpenAI execution). Use when the user wants the Agent to execute commands, build/test code, manage files, or operate on a workspace. Always pair with sandboxing — `allowed`, `blocked`, `ignore`, or `readonly`.

ag2ai Bundle

Add a custom Python tool to an AG2 beta `Agent` using the `@tool` decorator. Use when the user wants to give an Agent a new capability backed by Python code (API calls, DB queries, computations, file ops). Covers sync and async tools, parameter typing, Pydantic schema customisation, returning typed `Input` / `ToolResult` (text / data / images / binary), `final=True` early-exit, and dependency injection via `Context` / `Inject` / `Variable` / `Depends`.

ag2ai Skill

Send images, audio, video, or documents into an AG2 beta `Agent` alongside text. Pass `ImageInput`, `AudioInput`, `VideoInput`, or `DocumentInput` as positional args to `agent.ask(...)`. Use when the user wants the agent to process non-text input — describe a photo, transcribe audio, summarise a PDF, analyse a video. Covers per-provider support matrix, the four ways to source data (URL / path / bytes / file_id), Gemini-specific YouTube + media-resolution + clipping, OpenAI image-detail, Anthropic prompt-caching on attachments, and `FilesAPI` for upload lifecycle.

ag2ai Bundle

Get a typed Python value back from an AG2 beta `Agent` instead of free text. Pass `response_schema=` (a Pydantic model, dataclass, primitive, union, `ResponseSchema`, or `@response_schema` validator) and read the parsed result via `await reply.content()`. Use when the user wants validated structured output, classification, extraction, or scoring. Covers `ResponseSchema`, `@response_schema`, `PromptedSchema` (for providers without native structured output), per-turn override, validation retries, and primitive embedding.

ag2ai Bundle

Wire AG2 beta's shipped tools into an `Agent` — both provider-native server-side tools (web search, web fetch, code execution, MCP, image generation, memory) and locally-executed common toolkits (filesystem, DuckDuckGo, Exa, Tavily, skills). Use when the user wants capabilities AG2 already ships rather than writing custom Python. For shell commands see `ag2-shell-tool`; for custom Python tools see `ag2-add-custom-tool`.

ag2ai Bundle

Delegate work from one AG2 beta `Agent` to another. Two patterns — auto-injected `run_subtask` / `run_subtasks(parallel=True)` (opt in via `tasks=TaskConfig(...)`) for self-delegation and parallel fan-out, and `Agent.as_tool()` for named delegates between distinct agents. Use when one coordinator should spawn sub-tasks, fan out concurrent work, or hand off to a specialist agent. Covers context flow, recursion safety, and `persistent_stream` for sub-task history.

ag2ai Bundle

Persist agent state across runs, shape what the LLM sees per turn, and cap history to fit a context window. Covers `KnowledgeStore` (memory / sqlite / disk / redis), `KnowledgeConfig` (`store=`, `compact=`, `aggregate=`, `bootstrap=`), aggregation strategies (`WorkingMemoryAggregate`, `ConversationSummaryAggregate`), assembly policies (`WorkingMemoryPolicy`, `EpisodicMemoryPolicy`, `ConversationPolicy`, `SlidingWindowPolicy`, `TokenBudgetPolicy`, `AlertPolicy`), and compaction (`TailWindowCompact`, `SummarizeCompact`). Use when the user wants the agent to remember between conversations, manage long histories, or control prompt assembly.

ag2ai Bundle

Monitor an AG2 beta agent's stream — log events, detect repeated tool calls, track token spend, build trigger-driven observers, route observer alerts to the model, and halt on FATAL conditions. Covers `@observer(...)` (stateless), `BaseObserver` (stateful), built-ins (`TokenMonitor`, `LoopDetector`), `Watch` primitives (`EventWatch`, `CadenceWatch`, `DelayWatch`, `IntervalWatch`, `CronWatch`, `AllOf`, `AnyOf`, `Sequence`), `ObserverAlert` (`Severity.INFO/WARNING/CRITICAL/FATAL`), `AlertPolicy`, and `HaltEvent`. Use when the user wants observability, runtime safety guards, alerts, or batch/time-based reactive logic.

ionclaw-org Skill

Connect to external MCP servers to use their tools and read their resources. Use when the user wants to interact with a remote MCP-compatible service (e.g. another IonClaw instance, a database MCP server, a filesystem MCP server, or any third-party MCP server).

librefang Skill

Compose durable multi-step workflows with the workflow_create tool — step shape, agent binding, required skills, and the validation errors worth avoiding

vlabsai Bundle

Discover and install agent skills from the skills.sh ecosystem and local installed skills. Use when the user asks 'how do I do X', 'find a skill for X', 'is there a skill for...', 'can you do X' (where X is a specialized capability), or expresses interest in extending agent capabilities. Also use when the user wants to search for tools, workflows, or domain-specific help that might exist as an installable skill.

vlabsai Bundle

Agent-orchestrated research that takes input text and conducts parallel multi-agent investigation. Accepts research questions, briefs, or seed context. Spawns parallel agents for web research, local project analysis, and specialized investigation. Produces citation-backed markdown reports. Triggers: "deep research", "research X", "comprehensive analysis", "investigate", "compare X vs Y", "analyze trends". Do NOT use for simple lookups (1-2 searches), debugging, or document analysis without web research.

therapys Skill

Generate or refresh a repo's AGENTS.md (the source-of-truth agent context file) and its one-line CLAUDE.md pointer. Use when the user says 'update the claude.md', 'update AGENTS.md', 'analyze this project and update the context file', 'write an AGENTS.md', or after a stack/structure change makes the current file stale.

therapys Skill

Run the agent as an autonomous maintainer of a product — set it up once, then loop on small, verified, logged improvements while keeping the product deployable at all times. Use when the user says 'maintain this product by itself', 'run it on autopilot', 'set up an autonomous maintainer', 'keep improving this overnight', 'self-maintaining product loop', or wants the agent to continuously ship and verify changes until stopped.

armelhbobdad Bundle

Skill compilation specialist — the forge master. Use when the user asks to "talk to Ferris" or requests the "Skill Forge agent."

seasonedcc Skill

Write and run Vitest unit tests against a real throwaway SQLite database and end-to-end specs that drive the real MCP server over stdio. Use when writing or fixing a test, running test:unit, test:e2e, or test:seed-coverage, practising TDD, mutation-proving an assertion, adding fixtures, editing the E2E seed or its fake gateway feed, writing a seed demonstration, or working on the tool coverage gate or the dev-seed coverage gate.

seasonedcc Skill

Manage environment variables through the two-tier typed env pattern — a framework env and an app env, both built with make-typed-env and Zod. Use when adding or renaming an env var, editing app/framework/env.server.ts or app/env.server.ts, updating .env.example, or wiring configuration into framework, business, MCP, or ingest code.

seasonedcc Skill

Spawn subagents and dynamic workflows well — size each task to the context window (~33% of 1M target) and pick the right model tier. Use whenever delegating work to subagents, launching a Workflow, or deciding how to split a task across agents.

seasonedcc Skill

Build and extend the MCP stdio server in app/mcp/ — the product's entire user surface — keeping the tool surface equal to the business surface. Use when adding or changing a tool, wiring a new domain into the server, editing app/mcp/*, working on the parity test or its exemptions, naming a tool, or shaping a tool's input schema at the JSON boundary.

seasonedcc Skill

Write minimal, correct TypeScript — inference first, annotations only where removing them would lose safety, and no `any`. Use when adding types, declaring variables, writing a function signature, creating a type alias, deriving a type from a Zod schema, or reviewing code for redundant annotations and stray assertions.

seasonedcc Skill

Run delegated work reliably — writing subagent charters, verifying subagent claims, keeping judgment at the orchestrator tier, ledger discipline, recovering from interruptions, and shipping and merging lane PRs. Use when spawning subagents or workflows, coordinating parallel lanes, resuming after a session limit or compaction, or rebasing and merging a lane's PR. The subagents skill covers task sizing and model choice; this skill covers everything after the spawn.

seasonedcc Skill

Create, manage, and debug Claude Code Agent Skills. Use when creating new skills, debugging skill activation issues, writing SKILL.md files, managing skill structure, or learning about Claude Code skills. Helps with personal skills, project skills, YAML frontmatter, descriptions, and troubleshooting.

seasonedcc Skill

Organize domain logic in app/business/ by domain cohesion and transport independence. Use when creating a file in app/business/, adding a function to one, naming a business module, deciding whether logic belongs in business or in a transport (app/mcp/, app/ingest/), or choosing what a business file may import.

oozoofrog Bundle

Send requested questions, reviews, or files to ChatGPT Chat or Work, choose model and reasoning for the prompt's purpose and complexity, then return and process the completed response. Use for $gptplease or explicit ChatGPT consultation requests. Local skill explanation/editing or prompt drafting alone does not send a task.

oozoofrog Bundle

Run Jev Ultrafast browser tasks with TypeSafe action selection and Codex subscription text generation using gpt-5.6-luna. Use when the user requests this browser agent or its local inspector.

oozoofrog Bundle

Use the global Swift Intelligence MCP tools to navigate and inspect Swift code semantically. Apply for Swift definitions, references, protocol implementations, compiler-resolved types, document symbols, workspace symbols, and diagnostics; do not use for plain text or resource-file searches.

langwatch Skill

Version and manage your agent's prompts with LangWatch Prompts CLI. Use for both onboarding (set up prompt versioning for an entire codebase) and targeted operations (version a specific prompt, create a new prompt version). Supports Python and TypeScript.

langwatch Skill

Take your AI agent to the next level with full LangWatch integration. Adds tracing, prompt versioning, evaluation experiments, and simulation tests in one go. Use when the user wants comprehensive observability, testing, and prompt management for their agent.