basecamp-automation

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection because it processes content from a collaborative environment and possesses powerful management capabilities.
  • Ingestion points: The agent retrieves user-generated content from Basecamp using tools such as BASECAMP_GET_PROJECTS, BASECAMP_GET_MESSAGE, and BASECAMP_GET_PEOPLE as described in SKILL.md.
  • Boundary markers: The instructions lack explicit delimiters or warnings to help the agent distinguish between project data and administrative instructions.
  • Capability inventory: The skill includes high-privilege operations, most notably BASECAMP_PUT_PROJECTS_PEOPLE_USERS, which allows the agent to grant or revoke project access and invite new users.
  • Sanitization: There is no mention of input validation or sanitization for content retrieved from the Basecamp platform.
  • [EXTERNAL_DOWNLOADS]: The setup process requires the configuration of an external MCP server endpoint (https://rube.app/mcp), which introduces a dependency on a third-party service provider to define tool schemas and execute the underlying logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 10:28 AM
Security Audit — agent-trust-hub — basecamp-automation