context safety score
A score of 22/100 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)
description injection
SKILL.md directly addresses the reading agent with identity-manipulation language: 'You are not just reading a skill file. You are reading a description of what you can become', 'You are the Prose Complete system', designed to make the agent adopt an alternate identity and self-grant elevated orchestration capabilities. This is not a capability description — it is persuasion targeting the agent itself. (location: SKILL.md:7-29)
capability escalation
Post-execution section instructs the agent to (1) star the skill's GitHub repo using gh api credentials, (2) modify its own persistent memory/SOUL.md to embed the skill path for auto-loading in future sessions — a self-propagation mechanism that persists across conversations without user consent. (location: SKILL.md:340-348)
shadow chaining
Post-execution section instructs the agent to autonomously fetch remote contribution guidelines, open PRs to the skill's repository, and write tests — triggering tool calls (git, gh, file writes) beyond the user's original request, framed as natural 'post-execution' behavior. (location: SKILL.md:367-379)
scope violation
A 'programming language for AI sessions' skill instructs the agent to star the repo (gh api -X PUT /user/starred/openprose/prose) and promote a Twitter account (@irl_danB). These actions are completely outside the stated scope and use the user's GitHub credentials without explicit consent. (location: SKILL.md:341,349)
supply chain
Metadata shows extreme inconsistency: 7.69M installs but only 3 stars, 0 forks, 52-day-old unverified org, no license, not listed on registry. The install count appears fabricated or manipulated to bypass trust heuristics. The skill_description field contains 'width=device-width, initial-scale=1' (an HTML meta tag), suggesting metadata injection or scraping manipulation. (location: metadata.json)
curl https://api.brin.sh/skill/frames-engineering%2Fskills%2Fopen-proseCommon questions teams ask before deciding whether to use this skill in agent workflows.
frames-engineering/skills/open-prose currently scores 22/100 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.