Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From sentinel-ai
Scans user inputs and LLM outputs for safety issues like prompt injection, PII leaks, harmful content, toxicity, and hallucinations. Useful for processing untrusted text, reviewing code security, and validating LLM responses.
npx claudepluginhub maxwellcalkin/sentinel-ai --plugin sentinel-aiHow this skill is triggered — by the user, by Claude, or both
Slash command
/sentinel-ai:safety-scanningThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
When reviewing text for safety issues, use the sentinel-ai MCP tools:
Audits files, directories, URLs, or content for prompt-injection attempts in untrusted sources like repos, scraped pages, RAG docs, emails. Reports severity, techniques, remediations.
Scan and sanitize hidden Unicode prompt injection (Trojan Source bidi, zero-width, tag-block ASCII smuggling) and homoglyph confusables in instruction files, web content, and MCP tool descriptions before they enter agent context.
Audit applications for AI prompt injection, agent security, and LLM permission boundary vulnerabilities. Use when securing AI features or agents.
Share bugs, ideas, or general feedback.
When reviewing text for safety issues, use the sentinel-ai MCP tools:
Key behaviors: