Results for “colorblind-safe”

Implements container security with image scanning, runtime protection, image signing, and security policies using tools like Falco, Trivy, and Notary.

Scan container images for vulnerabilities with Trivy and Grype, generate SBOMs, sign images with Cosign and Sigstore, configure registry access controls, and enforce security gates in CI/CD pipelines.

Deploy Llama Guard, NeMo Guardrails, and LLM Guard as runtime input/output scanners to block jailbreaks, prompt injection, and toxic content in production LLM applications.

Manage environment variables securely by validating schemas, masking sensitive values, and preventing secret exposure in logs or agent context.

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.

Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.

Reviews authorized diffs for auth, input, filesystem, network, secrets, parsers, injection, CI/CD, and supply-chain regressions.

Chief Security Officer mode. Infrastructure-first security audit: secrets archaeology, dependency supply chain, CI/CD pipeline security, LLM/AI security, skill supply chain scanning, plus OWASP Top 10, STRIDE threat modeling, and active verification. Two modes: daily (zero-noise, 8/10 confidence gate) and comprehensive (monthly deep scan, 2/10 bar). Trend tracking across audit runs. Use when: "security audit", "threat model", "pentest review", "OWASP", "CSO review". (gstack) Voice triggers (speech-to-text aliases): "see-so", "see so", "security review", "security check", "vulnerability scan", "run security".

Performs a full-stack security posture assessment with 0-100 risk scoring, scanning dependencies, code patterns, configuration, and data handling, then produces a prioritized report and routes to specialized skills.