coaching-session
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by collecting unvalidated user input and writing it into persistent session plan files.
- Ingestion points: User input collected throughout Steps 2-7, including KPI definitions, session objectives, and reflection questions in SKILL.md.
- Boundary markers: The skill lacks delimiters or protective instructions to prevent embedded user content from influencing downstream processes that read the generated JSON or Markdown files.
- Capability inventory: The skill utilizes file-writing capabilities to save data to the local workspace.
- Sanitization: No sanitization, escaping, or validation logic is applied to the user-provided strings before file generation.
- [PROMPT_INJECTION]: The skill constructs file paths using a user-controlled
{slug}variable. Without proper validation or sanitization of this variable, an attacker could attempt path traversal (e.g., using../sequences) to write files to unintended directories on the system.
Audit Metadata