worker-dispatch
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to the processing of external data and configuration files.
- Ingestion points: The instructions require the agent to read
.agent-team/rules/index.mdand load "referenced task artifacts" during the dispatch process (SKILL.md). - Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined for the ingested files or artifacts.
- Capability inventory: The skill is authorized to use CLI tools (
agent-team worker open,agent-team worker status,agent-team reply) to manage worker state and communications. - Sanitization: There is no evidence of validation, escaping, or filtering of content derived from the task artifacts or rule files before processing.
Audit Metadata