specialist-support-developer-advocate
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses direct interpolation of untrusted user data into the system-level prompt for a spawned sub-agent, which is a common vector for indirect prompt injection attacks.
- Ingestion points: The variable
{the user's question}inSKILL.mdcaptures external input. - Boundary markers: Absent. The user input is appended directly to the end of the prompt without delimiters or instructions to the sub-agent to treat the input as data rather than instructions.
- Capability inventory: The skill utilizes the
Agenttool to spawn sub-agents. TheDeveloper-Advocatesub-agent is instructed to 'implement developer experience improvements', which could involve file system or code-related actions depending on available environment tools. - Sanitization: Absent. There is no evidence of escaping, filtering, or validation performed on the user's input before it is passed to the sub-agent.
Audit Metadata