The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes uv run to execute local Python scripts for configuration management and financial calculations. This includes the bundled scripts/founder_numbers.py script for runway and cap-table modeling, as well as environment scripts like resolve_customization.py and resolve_config.py located in the project root.
[EXTERNAL_DOWNLOADS]: The skill contains instructions to download and install additional components using npx skills add. These downloads are directed to the bmad-code-org GitHub organization, which is the official repository of the skill's author. Per security policy regarding vendor resources, these are documented as standard functionality.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, such as investor emails, customer interview notes, and call transcripts, creating a vulnerability to indirect prompt injection.
Ingestion points: Data enters the agent's context through files stored in the {record_folder}/raw/ directory, processed via the intake module (referenced in references/intake.md).
Boundary markers: The skill lacks explicit instructions for the agent to use delimiters or ignore embedded natural language instructions within the raw ingested text.
Capability inventory: The agent has access to file system write operations, local script execution via uv run, and the ability to invoke other skills that may perform network operations.
Sanitization: No specific sanitization or filtering logic is implemented for the content of processed emails or meeting notes.

bmad-founder-agent-advisor