procstats-review

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: This skill processes external data (SQL query results) provided by the user, representing a potential surface for indirect prompt injection. This behavior is intrinsic to the skill's primary diagnostic purpose and is not associated with any high-risk capabilities.
  • Ingestion points: User-pasted result grids from SQL Server report queries (SKILL.md).
  • Boundary markers: None; the skill does not explicitly use delimiters to separate the untrusted SQL data from instructions.
  • Capability inventory: The skill recommends additional diagnostic actions using other agent skills such as /sqlplan-review or /sqlwait-review.
  • Sanitization: None; the text is processed as provided by the user.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 11:13 AM