workflow-deployment

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by design, as it accepts a user's prompt as the primary context for a complex deployment workflow.
  • Ingestion points: Untrusted user data enters the agent context via the user's prompt passed to the /workflow-start deployment command in SKILL.md.
  • Boundary markers: The skill does not define any delimiters or explicit 'ignore embedded instructions' warnings to prevent the agent from following instructions hidden within the user context.
  • Capability inventory: The workflow triggers highly capable tools including /code (code generation), /test (execution), and /sre-review (infrastructure modification), which could be manipulated by a malicious prompt.
  • Sanitization: There is no evidence of input validation, escaping, or filtering of the external content before it is processed by the downstream workflow commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 07:59 PM