The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted user input regarding business value and customer segments without explicit boundary markers or sanitization instructions, creating a surface for indirect prompt injection.
Ingestion points: User responses to the coaching prompts in SKILL.md, where the user is asked to describe outcomes and customer metrics.
Boundary markers: The instructions lack explicit delimiters or instructions for the agent to ignore potentially malicious instructions embedded within the user-provided data.
Capability inventory: The skill logic is purely informational and does not invoke external tools, shell commands, or network operations, which limits the potential impact of an injection.
Sanitization: No explicit sanitization or filtering logic is defined for the ingested user content.
[SAFE]: The skill correctly references external sources and case studies from established academic, news, and technology platforms.
The skill identifies and links to documentation from well-known services and organizations, including OpenAI, Anthropic, Google, and JSTOR.
The skill references vendor-owned infrastructure at deciqai.com and github.com/deciqai for metadata and updates, which is standard behavior.

pricing-strategy