alterlab-nmc-portfolio-curator
Pass
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill's workflow involves ingesting untrusted data through web searches and the reading of project-specific files (Workflow section in SKILL.md). It lacks explicit boundary markers or instructions to sanitize this content, creating a vulnerability to indirect prompt injection where instructions hidden in external data could influence the agent's behavior. This risk is inherent to the autonomous execution mode and the tool's capability to read and write files. Mandatory Evidence Chain: 1. Ingestion points: Web search and local project files (Step 1: Audit & Inventory); 2. Boundary markers: Absent; 3. Capability inventory: Web search, file read, and file write (e.g., {project}-portfolio-plan.md); 4. Sanitization: Absent.
Audit Metadata