status-update
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize potentially untrusted data from the agent's execution environment, such as command outputs, worker states, and conversation history.
- Ingestion points: SKILL.md specifies reading the task's plan, recent turns, active worker states, and running command output.
- Boundary markers: Absent. The instructions do not specify using delimiters or warnings to ignore embedded instructions within the source material.
- Capability inventory: While the skill itself is instructed to be read-only, it operates within an agent environment that likely has broad file and command execution permissions.
- Sanitization: Absent. There is no explicit instruction to sanitize or filter content retrieved from live sources before summarization.
Audit Metadata