Splice for AI agents
Splice is a sample and sound library for music producers, who search its catalog, build stacks of sounds, and download the ones they use in a track. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Splice.
Search the sound catalog
Describe a sound to find matching samples in the catalog before adding one to a project.
Build a stack for a project
Create a stack of selected sounds for a track, grouping them for a producer to review.
Confirm before spending download credits
Check a sound's details before downloading it, since each download draws from a limited pool of credits.
6 Splice actions, graded by risk.
Every Splice action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
2Look things up. Allowed by default.
- splice.download_assetDownload asset
- splice.describe_a_soundDescribe a sound
write
4Create and change things. Allow, or ask the user first.
- splice.create_stackCreate stack
- splice.update_stackUpdate stack
- splice.share_stackShare stack
- splice.prompt_to_stackPrompt to stack
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
Splice in three steps.
- 01Your users connect SpliceThey sign in to Splice on Arc0 Connect, under your brand, and approve the access you ask for.
- 02You set the rulesReads run, and writes like “create stack” can wait for the user to approve.
- 03Any agent can actYour agent calls Splice through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('splice', { read: 'allow', write: 'ask', // create_stack destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Splice connects.
Splice runs its own MCP server behind OAuth. Arc0 registers the client, users approve access on Arc0 Connect, and your agent reaches Splice through the same endpoint, policies and audit log as every other app.
The same Splice connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- OAuth 2.0 (MCP)
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Splice from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Splice and Arc0, answered.
Can I use Splice with Claude, ChatGPT or Cursor?
Yes. Connect Splice to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Splice actions you allow.
How do users connect Splice?
Splice runs its own MCP server behind OAuth. Arc0 registers the client, users approve access on Arc0 Connect, and your agent reaches Splice through the same endpoint, policies and audit log as every other app.
Which Splice actions can my agent take?
6 in total: 2 read, 4 write and 0 destructive, such as “create stack”. Your policies decide which of them each agent may call.
Can I make my agent read-only in Splice?
Yes. Allow read actions and deny writes in the Splice policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call Splice too?
Yes. The same Splice connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Splice into your agent.
Your users connect Splice once, under your brand. Your agent gets 6 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan