iblai-api-apply
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill manages and processes untrusted data provided by applicants and form managers, which could potentially contain malicious instructions intended to influence the agent's behavior.
- Ingestion points: Data enters the system through
responsesin application forms anddecision_templatesused for automated messaging as described in the form schema and reviewer transition sections of SKILL.md. - Boundary markers: The skill documentation does not provide instructions for the agent to use delimiters or ignore embedded instructions when interpreting user-supplied responses or templates.
- Capability inventory: The skill allows the agent to perform network requests (curl) and modify data (POST/PATCH/DELETE) on the IBLAI platform, including sensitive operations like deciding on applications and assigning courses.
- Sanitization: Placeholder validation is mentioned for decision templates, but no specific sanitization or escaping mechanism is described for handling these inputs within a natural language context.
Audit Metadata