The Agent Skills Directory

[SAFE]: The skill architecture prioritizes safety through a tiered transformation model and automated testing checkpoints. No malicious patterns, data exfiltration, or persistence mechanisms were detected.
[COMMAND_EXECUTION]: Employs the Bash tool to execute aptos move test and aptos move test --coverage. These are standard, well-known commands used to verify the correctness of the Move smart contracts being modernized and do not involve untrusted remote sources.
[PROMPT_INJECTION]: The skill processes untrusted Move source code, presenting a surface for indirect prompt injection. (1) Ingestion points: The skill uses the Read tool to ingest .move files from the sources/ directory. (2) Boundary markers: No specific delimiters are implemented during code ingestion. (3) Capability inventory: The skill possesses the Write, Edit, and Bash tools. (4) Sanitization: Transformations are constrained by predefined rules in references/detection-rules.md and are validated against the project's existing test suite.

modernize-move