refactor-ops
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes project source code as untrusted input. Maliciously crafted content within the code, such as comments or string literals, could attempt to manipulate the agent's behavior during refactoring tasks.\n
- Ingestion points: The skill instructs the agent to read local files using Read, Glob, and Grep tools (specified in SKILL.md).\n
- Boundary markers: Absent; the instructions in SKILL.md and reference files do not define delimiters to isolate code content from instructions.\n
- Capability inventory: The agent is empowered by SKILL.md to use Write, Edit, and Bash tools to modify files and execute analysis tools.\n
- Sanitization: Absent; the skill lacks explicit mechanisms to filter or sanitize the content of ingested source code.
Audit Metadata