The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its codebase analysis feature which ingests external data. \n- Ingestion points: The MigrationAnalyzer._analyze_source function in resources/implementation-playbook.md reads the content of every file in the source directory. \n- Boundary markers: No specific delimiters or warnings are used to prevent the agent from following instructions found within the source files. \n- Capability inventory: The skill is capable of file system access, database interaction, and network communication through its generated scripts. \n- Sanitization: Source code content is read and processed without filtering or sanitization. \n- [COMMAND_EXECUTION]: The skill generates templates for CLI tools and automated migration scripts that interact with the file system and execute migration tasks. \n- [EXTERNAL_DOWNLOADS]: The Progress Monitoring section in resources/implementation-playbook.md includes a dashboard template that fetches the Chart.js library from https://cdn.jsdelivr.net/npm/chart.js. \n- [DATA_EXFILTRATION]: The skill provides templates for API resolvers and database migration scripts that perform network requests and data transfers between systems.

framework-migration-code-migrate