flow-phd-application
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a high-level orchestrator and routing logic for a set of external tools. It contains no executable scripts, system commands, or network-active code.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-provided materials (CVs, transcripts, GitHub profiles) which constitutes an attack surface for indirect prompt injection. * Ingestion points: Stage 1 (
apply-profile-reader) explicitly instructs the agent to ingest external documents and research writeups. * Boundary markers: The instructions do not prescribe specific delimiters or "ignore embedded instructions" wrappers for the ingested content. * Capability inventory: This specific file contains no subprocess, network, or file-system-writing capabilities, although it coordinates skills that may possess them. * Sanitization: No sanitization or validation logic is present for the external inputs. - [NO_CODE]: No code blocks or scripts are included in the skill definition.
Audit Metadata