The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions direct the agent to execute local Python scripts scripts/pmf_scorer.py and scripts/portfolio_analyzer.py for product-market fit assessment and portfolio analysis.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external business data (metrics, user feedback, and portfolio information) through external scripts, creating an attack surface for indirect prompt injection.\n
Ingestion points: External data processed by scripts/pmf_scorer.py and scripts/portfolio_analyzer.py during PMF and portfolio assessments.\n
Boundary markers: No explicit instructions or delimiters are provided in SKILL.md to prevent the agent from following instructions embedded within the processed data.\n
Capability inventory: Ability to execute Python scripts via the python command as documented in the Quick Start section.\n
Sanitization: No validation or sanitization mechanisms for the ingested business data are described in the provided instructions.

cpo-advisor