Catalog

Discover AI models, system tools, and community skills.

model
Gemini 2.5 Flash Lite

Google’s next-gen lightweight multimodal model.

googlegemini-2.5-flash-lite
model
Gemini 2.5 Flash

Balanced performance and speed for the 2.5 generation.

googlegemini-2.5-flash
model
Claude 4.5 Sonnet

Advanced reasoning and coding capabilities.

anthropicclaude-sonnet-4.5
model
Claude 4.5 Opus

Maximum intelligence for complex tasks.

anthropicclaude-opus-4.5
model
GPT-5.2

The latest evolution in the GPT series.

openaigpt-5.2
sandbox
OpenCode Agent

Autonomous AI Engineer (OpenCode CLI)

daytonadaytona-opencode
sandbox
Gemini CLI

Google Gemini development environment

daytonadaytona-gemini
sandbox
Dev Environment

Standard TypeScript/Python persistent environment

daytonadaytona-std
tool
net_scrape

Scrape a URL and return its content in Markdown format.

nativenet_scrape
tool
net_crawl

Start a crawler for a website. Returns a Job ID to track.

nativenet_crawl
tool
fs_write

Write text content to a file in the workspace storage. Automatically registers as a task artifact. Overwrites if exists.

nativefs_write
tool
fs_read

Read the text content of a file from workspace storage.

nativefs_read
tool
fs_list

List files in a directory.

nativefs_list
tool
fs_search

Semantic search for file content across the workspace. Uses vector embeddings to find relevant code or text.

nativefs_search
tool
sys.finish

Mark the task as completed and provide the final answer/artifacts. Always use this to end the conversation.

nativesys_finish
tool
sys.pause

Pause execution to ask the user for clarification or confirmation.

nativesys_pause
tool
agent_delegate

Delegate a complex task to a specialized sub-agent. This spawns a new agent process in a sub-folder.

nativeagent_delegate
tool
agent_ask

Ask a specific question to another agent or expert system (simulated). lightweight version of delegate.

nativeagent_ask
tool
compute_run_python

Execute Python code in a secure, isolated resource. Use this for calculation, data processing, or running algorithms.

nativecompute_run_python
tool
compute_run_node

Execute Node.js code/Javascript in a secure resource.

nativecompute_run_node
tool
compute_run_script

Execute a pre-defined script from a Skill. Safe way to run complex logic.

nativecompute_run_script
tool
mem_get

Retrieve a value from the workspace short-term memory (Key-Value).

nativemem_get
tool
mem_set

Store a value in short-term memory. Persists for the session.

nativemem_set
tool
mem_remember

Store a fact or text snippet into long-term semantic memory.

nativemem_remember
tool
mem_recall

Search long-term memory for relevant facts by semantic similarity.

nativemem_recall
tool
graph.add_node

Add a new node (task) to the execution graph. Can be linked to a parent or dependencies.

nativegraph.add_node
tool
graph.get_deps

Get dependencies for a specific task (upstream blockers) or children (downstream).

nativegraph.get_deps
tool
graph_add_dependency

Add a blocking dependency between two existing tasks (Parent blocks Child).

nativegraph_add_dependency
tool
db_write

Write a record to a specific named database (which is just a folder).

nativedb_write
tool
db_read

Read a record from a named database.

nativedb_read
tool
db_list

List keys in a named database.

nativedb_list
tool
hub_list_resources

Lists all available resources (agents, databases, compute) for the current workspace.

nativehub_list_resources
tool
hub_read_resource_thread

Reads the recent stream of a specific resource to see what it has been doing.

nativehub_read_resource_thread
tool
hub_send_message

Posts a message to a specific resource thread to delegate a task.

nativehub_send_message
tool
hub_get_status

Fetches the state JSON of a specific running resource.

nativehub_get_status