devrel-field-events
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources such as event briefs, interview responses, and guest lists during the event planning process. The agent is instructed to store task and project context in a local file named devrel-context.md. The instructions lack explicit boundary markers or sanitization procedures for this ingested data, which could allow malicious instructions hidden within a guest list or event brief to influence the agent's behavior or manipulate the contents of the memory file.
Audit Metadata