Runtime security guardrails for DeepSeek Harness: source-backed pre-install vetting, dangerous tool-call blocking, output-leak auditing, and tamper-evident HMAC-chained logs.
-
Updated
Aug 23, 2026 - JavaScript
#
guardrail
Here are 9 public repositories matching this topic...
Hard spend caps and a kill-switch for the non-LLM actions your AI coding agent takes. A drop-in Claude Code hook.
cli aws circuit-breaker ai-agents kill-switch finops guardrail cloud-cost claude-code agent-safety spend-cap
-
Updated
Aug 16, 2026 - JavaScript
Make coding agents follow proper procedure - config-driven task workflow for Claude Code + Codex + Linear + GitHub, with a git guardrail
-
Updated
Jul 13, 2026 - JavaScript
A policy hook for Claude Code
nodejs cli hooks security policy developer-tools ai-safety ai-agents guardrail prompt-injection agent-security claude-code
-
Updated
Jul 13, 2026 - JavaScript
Pre-publish compliance check for AI-generated social & ad content across Meta, TikTok, LinkedIn, Google Ads, YouTube, X, Snapchat, Pinterest.
social-media meta mcp compliance ai-agents google-ads brand-safety content-moderation tiktok guardrail ai-content llm-tools model-context-protocol ad-compliance advertising-policy
-
Updated
Aug 19, 2026 - JavaScript
唯稳律/Weiwen's Law — DSH 通用因果引擎：白箱可审计的 AI 安全中间件
-
Updated
Aug 31, 2026 - JavaScript
Pre-pay risk gate for autonomous payment agents — block prompt-injected destinations, intent mismatches & overcharges before your agent signs. Claude Agent SDK PreToolUse hook + any framework. Backed by 402Sentinel.
mcp ai-agents guardrail usdc prompt-injection agentkit x402 claude-agent-sdk agent-payments payment-firewall
-
Updated
Sep 1, 2026 - JavaScript
Completion verification gate for AI agents — re-fetch real world state with a probe before anything reports "done". Framework-agnostic, zero-dependency, no LLM.
-
Updated
Aug 18, 2026 - JavaScript
Hard-stop OpenCode model calls before you burn your Claude/Codex quota. Configurable 5h or weekly threshold, fail-safe blocking, and a live sidebar usage bar.
ai monitoring opencode rate-limiting tui openai quota budget cost-control codex claude guardrail usage-tracking llm anthropic opencode-plugin quota-monitoring
-
Updated
Jul 29, 2026 - JavaScript
Add this topic to your repo
To associate your repository with the guardrail topic, visit your repo's landing page and select "manage topics."