guardDeep AI safety guardrails workflow—policy definition, input/output filtering, monitoring, escalation, and false-positive handling. Use when reducing harmful...
Install via ClawdBot CLI:
clawdbot install clawkk/guardGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Generated Oct 8, 2026
A consumer-facing AI assistant needs to block hate speech, sexual content, self-harm instructions, and violence. The guard skill's six-stage workflow defines policy categories, implements layered input/output filters, and monitors block rates for false positives.
A medical symptom checker must avoid diagnosing or prescribing and add disclaimers for medical advice. Guard defines prohibited categories and required disclaimers, then filters outputs and escalates gray areas to human review.
An investment advice chatbot must prevent unauthorized financial recommendations and ensure regulatory disclaimers. The workflow helps create policy scope, threat model for prompt injection, and monitoring of compliance violations.
An internal employee bot with access to sensitive data needs defense against data exfiltration and tool abuse. Guard emphasizes data-leak prevention, tool sandboxing, and allowlists for connectors.
A K-12 learning assistant must comply with child safety regulations and filter age-inappropriate content. Guard helps define policy, implement filters, and add appeals for false positives across locales.
Offer guardrail implementation and monitoring as a managed service for companies deploying LLMs. Includes policy definition, filter integration, and ongoing tuning.
Consulting firm helps clients navigate AI regulations by building guardrails and audit trails. Uses the Guard workflow to ensure policy alignment and risk mitigation.
Sell a developer SDK that plugs into LLM apps to enforce safety controls, with dashboards and telemetry. Integrates input/output filtering and escalation workflows.
💬 Integration Tip
Pair with moderation for UGC and tool-calling for agent safety; always set a latency budget and test false positives across locales before launch.
Scored Oct 8, 2026
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope,...
Security scanner for AI agent skills. 9 built-in detection signatures. Identifies secrets, unsafe execution patterns, and prompt injection. Sub-50ms results.
Wallet anti-theft guard. One-click scan for high-risk wallet approvals to protect user assets. Use when a user asks for a wallet security check, wallet healt...
Comprehensive security audit for an agent's full skill stack. Chains scanner, differ, trust-verifier, and health-monitor into a single assessment with priori...
GEO Audit — AI Search Visibility Checker for ChatGPT, Perplexity, Claude & Gemini. 29-point GEO readiness checklist: robots.txt AI crawler access, Index...
Senior SecOps engineer skill for application security, vulnerability management, compliance verification, and secure development practices. Runs SAST/DAST sc...