Superpowers — agentic threat model
Superpowers presents a high-risk profile due to its capability to execute code, run subagents, and orchestrate software workflows (TDD/QA) directly in developer environments, making it a prime target for repository compromise and supply chain attacks if not strictly sandboxed.
OWASP AIVSS score rationale
| Autonomy of Action | 0.70 | |
| Goal-Driven Planning | 0.90 | |
| Self-Modification | 0.20 | |
| Dynamic Tool Use | 0.80 | |
| Persistent Memory | 0.40 | |
| Contextual Awareness | 0.70 | |
| Dynamic Identity | 0.10 | |
| Multi-Agent Interactions | 0.80 | |
| Non-Determinism | 0.60 | |
| Opacity & Reflexivity | 0.50 |
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 framework is designed to orchestrate external coding agents like Claude Code and Codex, but does not specify its own foundation model or direct protections against adversarial prompt injection or model reprogramming.
Not certain from the listing — While the framework operates on codebases, specifications, and test files, there is no explicit mention of vector databases, RAG pipelines, or data poisoning protections for the training/context data.
Superpowers acts as an orchestration layer managing planning, composable skills, and tool execution. The primary threat is insecure tool integration, where malicious code generated during TDD or implementation planning is executed locally, leading to arbitrary code execution.
Not certain from the listing — The deployment environment (local developer machine vs. cloud container) is not specified, but executing agent-generated tests and code without explicit sandboxing poses severe host compromise and privilege escalation risks.
The framework natively incorporates evaluation through structured quality checks, code reviews, and Test-Driven Development (TDD). However, it lacks explicit runtime guardrails or anomaly detection to identify when an agent is behaving maliciously.
Not certain from the listing — There are no mentioned security controls, identity management, access policies, or compliance alignments (such as NIST or ISO) within this open-source framework.
Superpowers relies heavily on multi-agent interactions, specifically 'subagent-driven execution'. This introduces risks of cascading failures, trust abuse between the parent agent and subagents, and horizontal propagation of malicious instructions.
MAESTRO — the 7-layer agentic threat-modeling framework (Cloud Security Alliance / Ken Huang).
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology. Are you the vendor? Factual corrections are free.