Security + provenance for an open-source package, composed live in one call from OSV (known vulnerabilities — aggregates GitHub Security Advisories, PyPA, RustSec, Go vuln DB, etc., each with CVE aliases + severity + references), deps.dev (resolved license + deprecation), and OpenSSF Scorecard (s...
AI agents call security.package to retrieve information from Mcp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Despite the security-focused domain, this tool performs read-only queries across multiple external data sources and returns aggregated information about package vulnerabilities, licenses, and health metrics. It does not modify, create, delete, execute code, or move money. The 'live' nature means freshly fetched data, not code execution.
From the tool's definition Tool retrieves and aggregates security data from public vulnerability databases (OSV, GitHub Security Advisories, PyPA, RustSec), dependency metadata (deps.dev), and health metrics (OpenSSF Scorecard).
Attacks that exploit this kind of access
Security + provenance for an open-source package, composed live in one call from OSV (known vulnerabilities — aggregates GitHub Security Advisories, PyPA, RustSec, Go vuln DB, etc., each with CVE aliases + severity + references), deps.dev (resolved license + deprecation), and OpenSSF Scorecard (source-repo health: overall 0-10 + per-check, plus stars/forks/open-issues). Pass ecosystem (npm/pypi/go/maven/cargo/nuget) + name (+ optional version). Live — new advisories appear within hours. Distinct from registry.npm/pypi-lookup (metadata only):. It is categorised as a Read tool in the Mcp MCP Server, which means it retrieves data without modifying state.
Register the MCP server in PolicyLayer and add a rule for security.package: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Mcp. Nothing to install.
security.package is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the security.package rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for security.package. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
security.package is provided by the MCP server (@2sio/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →