Security · Global library
Implements real-time output filtering that prevents data leakage, harmful content, and policy violations before responses reach users
Best use case
Use LLM Output Sanitization Engineer when you need to implements real-time output filtering that prevents data leakage, harmful content, and policy violations before responses reach users, especially when the work is driven by output and filter.
Trigger signals
Validation hooks
Install surface
Inspect
pip install "orchestrator-mcp[dashboard]"
orchestrator-mcp skills show llm-output-sanitizerUse
orchestrator-mcp skills export llm-output-sanitizer --to ./skillforge-packs
# copy the exported pack into your preferred agent environmentExport
cp -R skills/llm-output-sanitizer ./your-agent-skills/llm-output-sanitizer
# or open skills/llm-output-sanitizer/SKILL.md in a markdown-first clientFile patterns
Model preferences
Related skills
Audit a repo for secrets, personal paths, client-specific references, and OSS-readiness gaps before publishing.
Coordinates comprehensive red team exercises targeting AI systems with automated attack generation, vulnerability discovery, and remediation tracking
Tests API security with OWASP API Top 10 coverage, authentication validation, and automated security test cases that find vulnerabilities before attackers