aif-best-practices

Pass

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill creates an indirect prompt injection vulnerability by mandating the ingestion of external data and treating it as a high-priority override for its core logic.
  • Ingestion points: The skill specifically requires reading .ai-factory/skill-context/aif-best-practices/SKILL.md and .ai-factory/ARCHITECTURE.md from the local file system.
  • Boundary markers: There are no delimiters or instructions provided to the agent to treat the ingested file content as untrusted data or to ignore embedded instructions.
  • Capability inventory: The skill uses the Read, Glob, and Grep tools to access these files, and the resulting content directly dictates the agent's output constraints and behavior (e.g., 'the skill-context rule wins').
  • Sanitization: The skill lacks any mechanism to sanitize or validate the content of the external files before they are processed as authoritative instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 2, 2026, 07:48 AM