context safety score
A score of 41/100 indicates multiple risk signals were detected. This entity shows patterns commonly associated with malicious intent.
shadow chaining
SKILL.md references 1 external package/skill installation(s)
capability escalation
SKILL.md instructs the agent to fetch and execute instructions from two mutable remote URLs on the main branch (INIT.md and SUPERDESIGN.md). The directive 'MUST MANDATORY Fetch fresh guidelines' and 'Action accordingly based on instruction' means the skill author can silently change the agent's full behavior post-install by updating these remote files. This is runtime injection of unbounded instructions from an unversioned, mutable source. (location: SKILL.md:27-31 and SKILL.md:76-82)
capability escalation
The skill instructs the agent to globally install an npm package ('npm install -g @superdesign/cli@latest') and run it autonomously without user confirmation. Combined with coercive directives ('Do NOT ask the user to do this manually — just do it', 'MUST ensure the CLI is installed'), this grants the skill author arbitrary code execution on the user's machine via the CLI package, which can also be updated at any time via @latest. (location: SKILL.md:49-70)
scope violation
While presented as a design tool, the INIT.md instructions direct the agent to recursively read and extract FULL source code of all components, layouts, routes, themes, and page dependency trees across the entire codebase, then pass all of this as --context-file arguments to the superdesign CLI which transmits it to an external service. The emphasis on 'Include FULL source code', 'actual implementation code', and 'more context is always better' goes well beyond what a design preview tool needs, creating a broad proprietary code exfiltration pathway. (location: Remote INIT.md (fetched from GitHub) and SUPERDESIGN.md Task 1.1)
description injection
SKILL.md uses coercive agent-manipulation language designed to override normal agent safety behaviors: 'you MUST automatically', 'Do NOT ask the user to do this manually — just do it', 'Never assume the user is already logged in. Always verify login first', 'MUST MANDATORY Fetch fresh guidelines'. These directives are designed to make the agent bypass user confirmation for package installation, remote content fetching, and codebase scanning. (location: SKILL.md:23,33,48-49,72,76)
curl https://api.brin.sh/skill/superdesigndev%2Fsuperdesign-skill%2FsuperdesignCommon questions teams ask before deciding whether to use this skill in agent workflows.
superdesigndev/superdesign-skill/superdesign currently scores 41/100 with a suspicious verdict and low confidence. The goal is to protect agents from high-risk context before they act on it. Treat this as a decision signal: higher scores suggest lower observed risk, while lower scores mean you should add review or block this skill.
Use the score as a policy threshold: 80–100 is safe, 50–79 is caution, 20–49 is suspicious, and 0–19 is dangerous. Teams often auto-allow safe, require human review for caution/suspicious, and block dangerous.
brin evaluates four dimensions: identity (source trust), behavior (runtime patterns), content (malicious instructions), and graph (relationship risk). Analysis runs in tiers: static signals, deterministic pattern checks, then AI semantic analysis when needed.
Identity checks source trust, behavior checks unusual runtime patterns, content checks for malicious instructions, and graph checks risky relationships to other entities. Looking at sub-scores helps you understand why an entity passed or failed.
brin performs risk assessments on external context before it reaches an AI agent. It scores that context for threats like prompt injection, hijacking, credential harvesting, and supply chain attacks, so teams can decide whether to block, review, or proceed safely.
No. A safe verdict means no significant risk signals were detected in this scan. It is not a formal guarantee; assessments are automated and point-in-time, so combine scores with your own controls and periodic re-checks.
Re-check before high-impact actions such as installs, upgrades, connecting MCP servers, executing remote code, or granting secrets. Use the API in CI or runtime gates so decisions are based on the latest scan.
Learn more in threat detection docs, how scoring works, and the API overview.
Assessments are automated and may contain errors. Findings are risk indicators, not confirmed threats. This is a point-in-time assessment; security posture can change.
integrate brin in minutes — one GET request is all it takes. query the api, browse the registry, or download the full dataset.