WorkPPT — agentic threat model
WorkPPT is a low-autonomy productivity agent focused on document processing and slide generation. Its primary security risks stem from processing untrusted user documents (PDFs) and potential data exfiltration of sensitive uploaded files, rather than autonomous agentic actions.
OWASP AIVSS score rationale
| Autonomy of Action | 0.20 | |
| Goal-Driven Planning | 0.20 | |
| Self-Modification | 0.00 | |
| Dynamic Tool Use | 0.40 | |
| Persistent Memory | 0.10 | |
| Contextual Awareness | 0.30 | |
| Dynamic Identity | 0.00 | |
| Multi-Agent Interactions | 0.00 | |
| Non-Determinism | 0.50 | |
| Opacity & Reflexivity | 0.30 |
Scored with the canonical OWASP AIVSS formula (AIVSS calculator reference); agentic risk factors estimated from the agent’s described capabilities.
MAESTRO 7-layer threat model
Per-layer threats for this agent. Layers tagged “not certain from listing” are general, caveated commentary where the public description didn’t pin that layer.
Not certain from the listing — the underlying LLM is not specified, but it is vulnerable to prompt injection that could alter slide content or document summaries, and potential data leakage if the model is trained on user inputs.
Not certain from the listing — the agent processes user-uploaded PDFs and text documents. Risks include data exfiltration of sensitive documents, lack of secure deletion, and potential parsing vulnerabilities in PDF processing libraries.
Not certain from the listing — orchestration details are omitted. Risks include insecure tool execution during PDF manipulation (merge/split/watermark) and prompt injection hijacking the document generation flow.
Not certain from the listing — hosting and sandboxing details are unknown. Since it processes and manipulates PDFs, a lack of sandboxing could allow malicious PDFs to exploit parsing vulnerabilities and compromise the host.
Not certain from the listing — no monitoring, logging, or guardrails are mentioned. Gaps here could lead to undetected data exfiltration or generation of inappropriate content.
Not certain from the listing — no compliance certifications (e.g., SOC2, GDPR) or explicit access controls are mentioned, posing compliance risks for users uploading proprietary or PII-heavy documents.
The listing indicates this is a standalone productivity tool with no multi-agent or marketplace integrations, making ecosystem-level cascading failures or A2A trust abuse highly unlikely.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).