The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill framework is designed to ingest and process external, potentially untrusted data—such as meeting transcripts, stakeholder notes, and PRDs—to derive software requirements and tickets. 1. Ingestion points: Data entering the context as described in help/help.md and templates in the ingest-templates/ directory. 2. Boundary markers: The system utilizes Markdown headers and specialized HTML comment markers () as delimiters. 3. Capability inventory: The framework manages agents with capabilities for code execution (bmad-build), repository manipulation, and file-system access. 4. Sanitization: No specific input sanitization or validation of the natural language data is described in the documentation.
[REMOTE_CODE_EXECUTION]: The migration configuration files (migration-1.toml and migration-2.toml) contain instructions to execute internal Python scripts (migrate_v6.py and tickets.py) using the uv run command to upgrade project structures.

bmod-method