redis
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- PROMPT_INJECTION (HIGH): The skill is vulnerable to indirect prompt injection via documentation ingestion.
- Ingestion points: External documentation is pulled into the agent context via
mcp__context7__query-docsas described in SKILL.md. - Boundary markers: Absent; the skill does not define delimiters to isolate external documentation from system instructions.
- Capability inventory: The skill grants access to
Bash,Write, andEdittools, allowing for system-level changes. - Sanitization: Absent; fetched content is processed without validation or filtering.
- COMMAND_EXECUTION (LOW): The skill explicitly allows the
Bashtool. - Evidence:
Bashis listed in theallowed-toolssection of the YAML frontmatter in SKILL.md. While standard for development, this capability increases the impact of potential injection attacks.
Recommendations
- AI detected serious security threats
Audit Metadata