The Agent Skills Directory

[NO_CODE]: The skill consists exclusively of Markdown instructions and structured templates. It does not contain any executable scripts (Python, Node.js, Shell), automation logic, or binary files.
[DATA_EXFILTRATION]: There are no network operations, URL references, or commands that could be used to transmit data to external servers. The skill operates entirely within the agent's text generation context.
[CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or sensitive credentials were detected. The skill does not request access to environment variables or local credential stores.
[PROMPT_INJECTION]: The instructions include behavioral constraints and rules for managing the conversation flow. These are standard instructional parameters for defining an agent's persona and logic; no patterns typical of jailbreaks or safety filter bypasses were identified.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided content (pitch decks and sales scripts). While this creates a surface for indirect prompt injection if the user provides malicious instructions disguised as data, the skill lacks any capabilities (such as shell execution or file system writes) that could be exploited via this vector. The risk is limited to influencing the AI's textual output.

sales-narrative