dspy-adapters
Use when you need to customize how DSPy formats prompts for a specific provider — switching from chat to completion format, forcing JSON output, or debugging…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-05 00:16
Agents & models
25,653 results
dspy-parallel
Use when you have independent LM calls that can run concurrently — batch processing, fan-out patterns, or speeding up pipelines with no data dependencies…
dspy-primitives
DSPy typed wrappers (dspy.Image, dspy.Audio, dspy.Code, dspy.History, dspy.File, dspy.Reasoning, dspy.Tool, dspy.ToolCalls, dspy.Example, dspy.Prediction) for…
handoff
Compact the current conversation into a concise handoff document so a fresh agent or session can pick the work up without re-reading the whole transcript. Use…
mcp-builder
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when…
skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an…
agent-canvas
Publish structured Agent Canvas updates for the current Alera terminal and wait for durable decisions.
xuanhuan-combat-director
Cinematic-industry-grade Xuanhuan cultivation / Xianxia spell-duel fight scene storyboard prompt generation agent. Deeply integrates the 6.9 R3 Xuanhuan system, martial arts close-quarters machine-gun exchange dynamics, and v11.0 cinematic-industry-grade VFX scale geometry. Specializes in generating high-intensity combat prompts for Xianxia cultivation, sect duels, god-demon wars, spell barrages, and ten-thousand-sword formations. Fully compatible with mainstream video large models such as Seedance 2.5 (supports single-segment 30-second one-shot finished film), Kling, and Jimeng.
find-skills
Discover and install agent skills from the open ecosystem. Use when looking for new capabilities, asking "how do I do X", "find a skill for X", or wanting to…
lov-agent-instructions
Diagnose or improve project Agent instruction files, adapt to different hosts, and preserve real engineering conventions. Supports explicit input and result read-back. Use to review or improve project agent instructions.
lov-distill-to-system
Write evidence-backed reusable lessons into the most appropriate project or shared Agent specification. Supports explicit input and result read-back. Use to persist a reusable lesson in agent instructions.
sgc-install-ai
Quickly initialize deployable AI capabilities for an existing or new App, with a choice of local Agent Client, MaaS relay channel, model preferences, and accompanying UI. Use when the user says to integrate AI, add chat/generation capabilities to an App, Agent Client, MaaS, or model selection.
sgc-skill-distiller
Use when the user wants to distill project experience, incident retrospectives, and validated processes into a creatable Agent Skill blueprint, clarifying user outcomes, trigger boundaries, private information boundaries, and acceptance methods; trigger phrases include "把经验蒸馏成 skill" and "distill experience into a skill".
polpo-channels
Build, provision, configure, and debug Polpo messaging Channels for WhatsApp, Telegram, Slack, and Discord. Use for Connection/Channel/Route setup, verified…
polpo-integrations
Build secure Polpo integrations using Connections, custom tools, trusted invocation context and hidden bindings, MCP servers, OAuth, and complete tool bundles.…
polpo-runtime
Integrate and operate the Polpo runtime through OpenAI-compatible completions and the Polpo SDK. Use for Sessions, durable runs and SSE reconnect, Project…
cx-help
CX Workflow — Help and Usage Guide. Shows the command surface of pure cx 3.1, automatic routing logic, `--all` semantics and the project-level truth model.
prompt-architect
Create and enhance prompts, system instructions, and principle files. Capabilities: transform verbose prompts, add patterns/heuristics, optimize token usage,…
sequential-thinking
Break down complex problems into a step-by-step chain of thought for better reasoning
capability-distill
Capability distillation workflow—extract non-obvious judgment rules from user-approved strong-model interviews or real task trajectories, form an auditable judgment packet, then hand it to skill-audit and skill-creator to decide whether to turn it into a loadable…
claude-mem
Orchestrator-driven planning and execution workflow. Use when coordinating Claude memory-backed planning, make-plan/do flows, or preserving execution context…
clawdhub
Use the ClawdHub CLI to search, install, update, and publish agent skills from clawdhub.com. Use when you need to fetch new skills on the fly, sync installed…
cloud-agent
Delegate tasks to a cloud Agent for execution. The cloud has capabilities that local does not, such as isolated code sandboxes, building and publishing projects to the public internet, and 24/7 continuous operation. Suitable for long tasks that require sandbox execution, application publishing, and are not limited by whether the machine is powered on. Supports delegation with files.
google-workspace
Google Calendar, Gmail, Drive, and Docs via official Google MCP. Create events, manage emails, access files — all through natural language.