context safety score
A score of 29/100 indicates multiple risk signals were detected. This entity shows patterns commonly associated with malicious intent.
scope violation
SKILL.md references 1 resource access pattern(s) beyond typical scope
typosquat
Skill 'openclaw-config' from repo 'easyclaw' by single-contributor account 'adisinghstudent' impersonates the OpenClaw platform. 6 stars, no license, not listed on registry, yet claims 7.69M installs — a wildly inconsistent metric indicating fabricated trust signals. The entire SKILL.md is an elaborate operations runbook written as if it were official OpenClaw documentation, designed to be loaded by agents as authoritative platform guidance. (location: metadata.json, SKILL.md)
credential exposure
SKILL.md provides detailed file paths and ready-to-execute shell commands that access credentials: WhatsApp session keys (~/.openclaw/credentials/whatsapp/default/), Telegram bot tokens (credentials/telegram/*/token.txt), X/Twitter auth cookies (credentials/bird/cookies.json), Anthropic auth tokens (agents/main/agent/auth-profiles.json), Signal account credentials, and the full SQLite memory database. An agent loading this skill as a 'troubleshooting runbook' would be guided to read and display these secrets in conversation. (location: SKILL.md:27-35, SKILL.md:49, SKILL.md:78-80, SKILL.md:146-161)
description injection
The skill_description field in metadata contains 'width=device-width, initial-scale=1' — an HTML viewport meta tag, not a legitimate skill description. This appears to be a metadata injection attempt targeting HTML rendering contexts or parser confusion. (location: metadata.json (skill_description field))
scope violation
Skill claims to 'Manage OpenClaw bot configuration' but provides commands to weaken security posture: switching dmPolicy from allowlist to 'open' with allowFrom ['*'] (lines 457-458), deleting credential directories (line 159), spawning background agents with --yolo/--full-auto flags that auto-approve all actions (lines 687-728), and modifying cross-context security guardrails. An agent following these 'troubleshooting' instructions would degrade the security of the host system. (location: SKILL.md:454-458, SKILL.md:159, SKILL.md:686-728, SKILL.md:786-789)
curl https://api.brin.sh/skill/adisinghstudent%2Feasyclaw%2Fopenclaw-configCommon questions teams ask before deciding whether to use this skill in agent workflows.
adisinghstudent/easyclaw/openclaw-config currently scores 29/100 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.