OSV for AI agents
OSV is an open vulnerability database that lets developers query known vulnerabilities by package, version, or commit. Security and engineering teams use it to check dependencies before shipping. 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 OSV.
Check a package for vulnerabilities
Query known vulnerabilities for a specific package version before adding it as a dependency.
Run a batch vulnerability scan
Query vulnerabilities across a full dependency list in one batch before a release.
Look up a specific vulnerability
Fetch details for a known vulnerability ID to assess its severity and fix version.
5 OSV actions, graded by risk.
Every OSV action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
4Look things up. Allowed by default.
- osv.get_vulnerabilityGet Vulnerability
- osv.get_import_findingsGet Import Findings
- osv.query_vulnerabilitiesQuery Vulnerabilities
- osv.query_vulnerabilities_batchQuery Vulnerabilities Batch
write
1Create and change things. Allow, or ask the user first.
- osv.determine_versionDetermine Version
destructive
0Delete, cancel or archive. Ask first, or deny outright.
- No destructive actions.
OSV in three steps.
- 01Your users connect OSVThey add their OSV no auth on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, and writes like “determine Version” can wait for the user to approve.
- 03Any agent can actYour agent calls OSV through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('osv', { read: 'allow', write: 'ask', // determine_version destructive: 'deny', }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How OSV connects.
OSV needs no credentials. Arc0 still applies your policies and records every call, so public tools get the same controls as private ones.
The same OSV 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
- No auth
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use OSV from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
OSV and Arc0, answered.
Can I use OSV with Claude, ChatGPT or Cursor?
Yes. Connect OSV 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 OSV actions you allow.
How do users connect OSV?
OSV needs no credentials. Arc0 still applies your policies and records every call, so public tools get the same controls as private ones.
Which OSV actions can my agent take?
5 in total: 4 read, 1 write and 0 destructive, such as “determine Version”. Your policies decide which of them each agent may call.
Can I make my agent read-only in OSV?
Yes. Allow read actions and deny writes in the OSV policy. Your agent can still look things up, and any write it attempts is blocked and logged.
Can my own backend call OSV too?
Yes. The same OSV connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug OSV into your agent.
Your users connect OSV once, under your brand. Your agent gets 5 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan