grackle-ai-plugin-core

Ecosystem: npm

Downloads: 395

Vulnerabilities from the last (months)
  • 0CriticalAffects 0 versions
  • 1HighAffects 1 version
  • 0MediumAffects 0 versions
  • 0ModerateAffects 0 versions
  • 0LowAffects 0 versions
Severity distribution (all time)
1vulnerability
  • Critical0 (0.0%)
  • High1 (100.0%)
  • Medium0 (0.0%)
  • Moderate0 (0.0%)
  • Low0 (0.0%)

Monthly breakdown

  • 1 vulnerabilities
    • Critical: 0
    • High: 1
    • Medium: 0
    • Moderate: 0
    • Low: 0

Vulnerabilities (1)

Known vulnerabilities affecting grackle-ai-plugin-core, based on data from OSV.

Vulnerabilities affecting grackle-ai-plugin-core
IDSeveritySummaryAffected versionsPublishedSource
GHSA-f9ff-5x35-7gfwHighGrackle: Fail-open authorization in the MCP tool layer lets scoped agents perform cross-task and cross-session mutations (IDOR)## Summary Authorization for scoped (agent) MCP callers is enforced **inline, per tool**, and is applied inconsistently — several mutating tools silently omit the ancestry/workspace check that their siblings perform. Because the MCP server authenticates all outbound gRPC with the full server API key and the backend gRPC handlers perform **no caller-based authorization**, the MCP tool layer is the *sole* authorization boundary. A malicious or prompt-injected scoped agent can therefore perform cross-task and cross-session operations it should not be allowed to (an IDOR / privilege-boundary bypa
  • <= 0.132.1
GitHub (opens in a new tab)
  • GHSA-f9ff-5x35-7gfwHigh
    Grackle: Fail-open authorization in the MCP tool layer lets scoped agents perform cross-task and cross-session mutations (IDOR)
    • <= 0.132.1

    ## Summary Authorization for scoped (agent) MCP callers is enforced **inline, per tool**, and is applied inconsistently — several mutating tools silently omit the ancestry/workspace check that their siblings perform. Because the MCP server authenticates all outbound gRPC with the full server API key and the backend gRPC handlers perform **no caller-based authorization**, the MCP tool layer is the *sole* authorization boundary. A malicious or prompt-injected scoped agent can therefore perform cross-task and cross-session operations it should not be allowed to (an IDOR / privilege-boundary bypa

Showing 1-1 of 1 vulnerabilities