62 tagged AI Safety, measured the same way as everything else here.
Browse within: ai-security 19ai-behavior-analysis 15ai-safety-research 15claude-code-plugin 15ai-skill-safety 14ai-governance 7Guardrails 6
microsoft/agent-governance-toolkit
Plugin Claude Code ✓ vendor
Bundles 2 commands, 1 MCP server · 32 tokens together
AGT governance hooks and MCP tools for Claude Code sessions.
Plugin Claude Code
Bundles 1 skill, 2 hooks, 1 plugin · 90 tokens together
Security controls for AI agents — deterministic policy enforcement, OWASP ASI10 scanning, and audit trails.
Plugin Claude Code
Bundles 1 skill, 1 command · 35 tokens together
Anti-hallucination research mode for Claude Code. Toggle on/off to enforce citation requirements, source grounding, and 'I don't know' behavior.
Plugin Claude Code
Bundles 1 skill, 1 MCP server · 78 tokens together
Runtime AI-threat screening for Claude: analyze prompts, scan tool calls and outputs, and vet documents before you act, with BLOCK / CHALLENGE / ALLOW verdicts. Paid feature — needs a Pro-plan-or-higher Guardian API key.
Plugin Claude Code
Bundles 2 skills · 112 tokens together
Add and manage evaluation results in Hugging Face model cards. Supports extracting eval tables from README content, importing scores from Artificial Analysis API, and running custom model evaluations with vLLM/lighteval/inspect-ai.
Plugin Claude Code
Pre-ingestion verification for epistemic quality in RAG systems. Ensures documents are properly qualified before entering knowledge bases.
FlyFission/nuclear-grade-context-engineering
Plugin Claude Code
Bundles 29 skills, 28 commands, 5 agents · 2,677 tokens together
Risk-graded context engineering for AI-assisted software work — a catalog of agent skills, paste-ready command prompts, and change-record templates that keep agent work questioned, controlled, and reviewable.
Plugin Claude Code
Bundles 1 skill, 6 commands, 1 hook · 193 tokens together
Pre-execution safety layer that blocks dangerous shell commands and credential file reads using pattern matching + LLM analysis. Fail-closed design.
Plugin Claude Code
Bundles 74 skills, 7 commands, 1 agent, 5 hooks, 2 MCP servers, 2 plugins · 6,303 tokens together
One 👎 becomes a hard rule the agent cannot bypass. Captures thumbs-down feedback, distills it into PreToolUse Pre-Action Checks, enforced across every future Claude Code session.
Plugin Claude Code
Plugin marketplace listing 28 plugins: ai-agents-hardened-skills, coding-hardened-skills, devops-hardened-skills, documents-hardened-skills, finance-hardened-skills.
Plugin Claude Code
Bundles 2 skills, 4 commands, 1 agent · 186 tokens together
EU AI Act compliance scanner and trust infrastructure for Python AI projects.
Plugin Claude Code
Bundles 1 skill, 5 commands, 1 MCP server · 52 tokens together
Real-time safety scanning for LLM interactions. Detects prompt injection, PII leaks, harmful content, toxicity, obfuscation, secrets, and dangerous tool calls — 600-case benchmark at 100% accuracy with sub-millisecond latency.
Plugin Claude Code
Bundles 7 skills, 8 hooks · 555 tokens together
Nio — Execution assurance and observability for autonomous AI agents. Dynamic guard, static scan, OTEL collector.
Plugin Claude Code
Bundles 2 skills, 5 hooks · 51 tokens together
Production safety hooks for autonomous Claude Code operation. Context monitoring, syntax checking, branch protection, activity logging, and more.
datumbrain/claude-code-privacy-guard
Plugin Claude Code
Privacy guard plugin for Claude Code - prevents PII and secrets from leaking into AI prompts.
Plugin Claude Code
Bundles 5 skills, 3 agents · 303 tokens together
Programmatic Agent Constraint Toolkit — governance infrastructure for AI coding agents.
Plugin Claude Code
Bundles 1 skill · 235 tokens together
Strip the sycophancy. Makes Claude honestly audit its own claimed-complete work against 17 named under-delivery patterns instead of polishing a summary over half-finished tasks.
Plugin Claude Code
Detects shortcut-taking behavior and guides Claude to follow procedures step by step.
bitflight-devops/hallucination-detector
Plugin Claude Code
Bundles 4 skills, 3 commands, 5 agents, 4 hooks, 1 MCP server · 708 tokens together
Stop-hook that audits assistant output for speculation, ungrounded causality, pseudo-quantification, and completeness overclaims — blocks completion to force evidence-first rewrites.
Plugin Claude Code
Bundles 8 skills, 11 hooks · 147 tokens together
Security enforcement layer for Claude Code. Blocks dangerous commands, audits every tool call, detects prompt injection.
Plugin Claude Code
Bundles 24 skills · 747 tokens together
Code Copilot Team: 24 skills, 14 agents, 14 commands and the SDD templates, plus hooks for file protection, git safety, auto-format, type verification, test-on-stop, context re-injection and notifications. Generated from the same sources as setup.sh; always-loaded rules, the launcher and the peer-review hooks need…
synaptiai/agent-capability-standard
Plugin Claude Code
Bundles 42 skills, 2 hooks · 1,496 tokens together
Grounded Agency: 36 atomic capabilities across 9 cognitive layers with typed contracts, safety-by-construction, and grounded reasoning for reliable AI agents.
Plugin Claude Code
Bundles 2 skills, 1 MCP server · 27 tokens together
5-layer cognitive memory for Claude Code — surprise filtering, semantic highlighting, Cortex classification, adaptive decay, cross-project learning. 30 MCP tools. 90.7% benchmark accuracy.
gg-mo/second-thought-agent-skill
Plugin Claude Code
Bundles 2 skills, 1 command, 1 hook · 100 tokens together
Decision checkpoint engine for agents that challenges drift, weak assumptions, and shallow convergence before action.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: