skills/djalmajr/skills/aim-query/Gen Agent Trust Hub

aim-query

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external memory instances through various MCP tools and API endpoints, creating a surface for indirect prompt injection.
  • Ingestion points: Data is pulled into the agent's context via tools like memory_query, memory_recent, memory_explore, and direct API access to /api/v1/.../pages/<path>.
  • Boundary markers: The skill instructions do not specify the use of delimiters or 'ignore' instructions to separate retrieved memory content from the agent's core operational logic.
  • Capability inventory: The agent uses the retrieved data to perform design decisions, status reports, and 'catch-up' summaries, potentially executing instructions embedded in the memory hits.
  • Sanitization: There is no evidence of content sanitization, validation, or escaping of the recalled markdown/prose content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 07:05 PM
Security Audit — agent-trust-hub — aim-query