The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill generates executable Python and JavaScript code at runtime based on the analysis of source codebases.
In resources/implementation-playbook.md, the Python2to3Migrator class uses the ast module and astor library to transform and generate new Python source code.
The create_migration_cli function generates a complete Python CLI script as a string.
The SQLToNoSQLMigrator class generates a database migration script for data transfer and transformation.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data (source codebases) which could contain malicious patterns.
Ingestion points: The MigrationAnalyzer in resources/implementation-playbook.md recursively reads all files from a user-provided source_path using Path.rglob('*').
Boundary markers: There are no explicit boundary markers or instructions to the model to ignore embedded malicious instructions within the source code files being analyzed.
Capability inventory: The skill provides functionality to generate and prompt the execution of migration scripts, CLI tools, and database operations.
Sanitization: There is no evidence of sanitization or validation of the content read from the source files before it is processed for migration analysis or script generation.
[EXTERNAL_DOWNLOADS]: The skill includes a dashboard template in resources/implementation-playbook.md that references an external script.
It fetches the chart.js library from jsDelivr's CDN (https://cdn.jsdelivr.net/npm/chart.js).

framework-migration-code-migrate