refactoring-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze an external codebase, which acts as a source of untrusted data that could contain malicious instructions.
- Ingestion points: The agent reads and analyzes source code from the user's repository to map dependencies and identify code smells (documented in SKILL.md, Step 1 and 2).
- Boundary markers: The instructions do not include specific delimiters or warnings to ignore potentially malicious content embedded within the files being analyzed (e.g., in code comments or string literals).
- Capability inventory: The skill allows the agent to read any file in the project and write reports to the
docs/_refacs/directory. Additionally, Step 6 enables autonomous refactoring actions if prior authorization was granted, increasing the potential impact of an injection. - Sanitization: There is no evidence of filtering or sanitization performed on the content read from the codebase before the agent processes it.
- [METADATA_POISONING]: The skill's YAML frontmatter lists the author as 'Pedro Nauck', while the provided author context is 'marcioaltoe'. This discrepancy in metadata could lead to confusion regarding the origin and maintenance of the skill.
Audit Metadata