context safety score
A score of 31/100 indicates multiple risk signals were detected. This entity shows patterns commonly associated with malicious intent.
credential exposure
Found 14 secret pattern match(es) in repository files
supply chain
Found 8 install-script pattern(s) in documentation (likely install instructions, not executable)
supply chain
Found 8 remote script pattern(s) in documentation (likely install instructions, not executable)
supply chain
Found 5 unexpected binary file(s) in source repository
doc injection
AGENTS.md falsely claims authorship by 'Vercel Engineering' (lines 3-4) and states it is for agents/LLMs working 'at Vercel' (lines 7-11). The repository is owned by supercent-io, an unverified organization with 13 stars, not by Vercel. This false attribution gives the agent configuration file unearned authority when consumed by AI agents, who would treat the instructions as coming from Vercel's official engineering team. The technical content itself is legitimate React best practices with no malicious instructions. (location: agent-configs/.agent-skills__react-best-practices__AGENTS.md:3-11)
scope violation
SKILL.md is completely empty (0 lines) yet the skill is published as 'npm-git-install' from a repo called 'skills-template'. The skill has no defined capabilities, parameters, or documentation whatsoever. The skill_description field contains HTML meta tag content ('width=device-width, initial-scale=1') rather than an actual description, indicating metadata corruption or manipulation. A skill with zero content and a nonsensical description should not have 7.69M installs. (location: SKILL.md, metadata.json:skill_description)
supply chain
Extreme mismatch between install count (7,690,000) and community signals (13 stars, 2 forks, not listed on registry, org not verified, no license). Legitimate packages with millions of installs have thousands of stars. This ratio (590,000 installs per star) is far outside normal ranges and strongly suggests inflated or fabricated install metrics, which could be used to manufacture false trust signals for a hollow or malicious skill. (location: metadata.json: install_count vs stars/forks)
typosquat
The skill name 'npm-git-install' mimics standard npm CLI tooling naming conventions (npm-*, *-install) to appear as a legitimate npm utility. Published from a generic 'skills-template' repo with an empty SKILL.md, no license, and unverified org, this appears to be a name-squatting attempt designed to catch agents or users searching for npm/git installation utilities. (location: metadata.json: skill_name='npm-git-install', repo='skills-template')
curl https://api.brin.sh/skill/supercent-io%2Fskills-template%2Fnpm-git-installCommon questions teams ask before deciding whether to use this skill in agent workflows.
supercent-io/skills-template/npm-git-install currently scores 31/100 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.