structural-refactoring
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The file
references/sequence.mdreferences the author's canonical repository athttps://github.com/lidessen/skills.gitfor principle lineage. This is a legitimate vendor-owned resource used to establish the skill's conceptual baseline. - [COMMAND_EXECUTION]: The skill requires the agent to execute verification surfaces, type checks, and tests (e.g., 'Run the current verification surface when practical') to validate that refactoring does not alter program behavior.
- [PROMPT_INJECTION]: The skill processes external source code and test files, which constitutes a surface for indirect prompt injection. Evidence Chain: 1. Ingestion points: Reads target source code, callers, and tests (
SKILL.md). 2. Boundary markers: The skill uses a structured grounding template but does not define explicit delimiters to isolate untrusted code content from instructions. 3. Capability inventory: Authorization to read files, perform structural edits, and execute local verification commands. 4. Sanitization: No specific filtering or escaping mechanisms are described for the ingested code content.
Audit Metadata