The Agent Skills Directory

[DYNAMIC_EXECUTION]: The compile.py script uses importlib.util to dynamically load and execute logic from a Python file located at ../founder-cfo/unit_economics.py. This technique of loading code from a computed path outside the skill's own directory represents a risk where a compromised file in a related skill could lead to unauthorized code execution within this skill's context.
[INDIRECT_PROMPT_INJECTION]: The skill functions as a compiler for data generated by other tools, creating a surface for indirect prompt injection.
Ingestion points: The skill reads content from multiple files in the founder/ directory, including idea.md, board.md, competitors.md, and others.
Boundary markers: There are no explicit boundary markers or instructions provided to the agent to treat the contents of these ingested files as untrusted data or to ignore any instructions embedded within them.
Capability inventory: The skill executes a local Python script (compile.py) and has the capability to write multiple markdown files to the filesystem.
Sanitization: The skill performs no sanitization or escaping on the content read from external files before including them in the final business plan output.

founder-plan