agent-cloud-architect
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses the
$ARGUMENTSplaceholder to ingest user-provided task descriptions directly into its operational context. - Ingestion points: The user input is appended directly to the end of the
SKILL.mdfile via the$ARGUMENTSvariable. - Boundary markers: The skill lacks explicit delimiters (such as XML tags or unique string markers) or instructions to the agent to ignore potentially malicious commands embedded within the user input.
- Capability inventory: As a
general-purposeagent, it likely has access to standard file system and command execution tools provided by the platform, which could be targeted by an injection attack. - Sanitization: There is no evidence of input validation, filtering, or escaping for the ingested content.
Audit Metadata