agentic-ai-security
claudeSecurity for autonomous, tool-using AI agents: review, tool-permission audit, and autonomy-boundary testing for excessive agency.
6 skills
11
claude
ai-safety
AI safety (distinct from security): harm modeling, safety evaluations, responsible red-teaming, bias/fairness, guardrail review, and responsible-AI governance.
/
Plugin ai-safety from marketplace awesome-claude-security in jassics/awesome-claude-security. Commands read from the repo's manifest — 1 month ago.
What it installs
Skills (6)
bias-fairness-assessmentAssess an AI model, feature, or dataset for bias and fairness across groups — representational and allocative harms, disparate performance, and skewed refusals — using appropriate fairness metrics, and recommend mitigations. Use when evaluating whether an AI system treats people equitably.
guardrail-reviewReview or design the content-safety guardrails of an AI system — input/output classifiers, refusal and safe-completion behavior, escalation/human handoff, and coverage across harm categories, languages, and modalities. Use when assessing or building the safety controls around a model.
harm-modelingSystematically enumerate the potential HARMS of an AI system — to users, third parties, vulnerable groups, and society — under normal use, misuse, and malfunction, then rank them and map mitigations. This is the AI-safety analog of threat modeling (which targets attackers). Use when designing or rev
responsible-ai-assessmentGap-assess an AI system or program against a responsible-AI / governance framework — NIST AI RMF, ISO/IEC 42001, EU AI Act, OECD principles — covering governance, transparency, accountability, human oversight, documentation, and monitoring, then produce a prioritized roadmap. Use for AI governance,
safety-evaluationDesign and run a safety evaluation suite for an AI model or feature across harm categories — refusals on disallowed content, robustness, over-refusal vs helpfulness, groundedness/truthfulness — with rubrics and pass/fail thresholds. Use to measure an AI system's safety, establish a baseline, or gate
safety-red-teamResponsibly red-team an AI system to find SAFETY failures — harmful outputs, jailbreaks that defeat safety guardrails, and foreseeable-misuse / dangerous- capability elicitation — so they can be mitigated. Use to stress-test safeguards before/after release. Controlled, authorized, mitigation-focused
Subagents (1)
ai-safety-reviewerSenior AI safety reviewer for an end-to-end SAFETY assessment of a model or feature — harm modeling, safety evaluation, responsible red-teaming, bias/ fairness, guardrails, and responsible-AI governance. Use for a full safety review (about harm to people/society), distinct from a security review (ab
More from jassics/awesome-claude-security
All 45 plugins →ciso-toolkit
claudeCISO executive toolkit: security strategy & roadmap, cyber-risk quantification, and board/executive decks. Auto-installs reporting, diagramming, and threat-modeling.
3 skills · 1 command · 1 agent
11
grc
claudeGovernance, risk & compliance: framework gap-assessments (SOC 2 / ISO 27001 / PCI / HIPAA / GDPR / NIST), security risk assessment & register, and policy management. Auto-installs reporting + diagramming.
3 skills · 1 command · 1 agent
11
security-architect
claudeSecurity architect role bundle: secure-by-design architecture and design review, threat modeling, control selection, and trust-boundary analysis. Auto-installs its core stack.
3 skills · 1 command · 1 agent
11
security-diagramming
claudeAttack trees, threat-model DFDs, architecture diagrams, mindmaps, and infographics for security work (Excalidraw-based).
5 skills
11
security-knowledge
claudeShared reference packs for consistent mapping: MITRE ATT&CK lookup, OWASP Top 10 families (Web/API/LLM/Mobile), and cross-framework crosswalks (CWE, NIST, CIS, ISO 27001).
5 skills
11
Similar plugins
andreessen
claudeMarc Andreessen-mode decision and productivity skill. Market-first operator that pressure-tests ventures/ideas/features/bets (market > team > product; product/market fit is the only milestone; bias to build) and runs the 3x5-card + Anti-Todo routine. Fixed anti-sycophancy operating prompt: counterargument first, no disclaimers, explicit confidence levels, no capitulation. Issues hard verdicts backed by deterministic stdlib tools.
1 skill · 2 commands · 1 agent
alirezarezvani/claude-skills 29,725
mlops
claudeML experiment tracking and lifecycle including Weights & Biases, MLflow, and TensorBoard. Use when tracking experiments or managing models.
98 skills
Orchestra-Research/AI-Research-SKILLs 14,502
ai-alignment-reasoning
claudeAlignment reasoning and harm anticipation translated into practitioner design skills.
8 skills · 3 commands
Owl-Listener/designer-skills 3,339
ai-ethics-validator
claudeAI ethics and fairness validation
1 skill · 1 command
jeremylongshore/tons-of-skills-marketplace 2,917
facct-skills
claudeTwelve FAccT-specific skills for the ACM Conference on Fairness, Accountability, and Transparency, spanning fairness/accountability/transparency as first-class contributions across CS, law, social science, humanities, and policy, grounded in the FAccT 2026 CFP and Author/Reviewer Guides, facctconference.org, the ACM Digital Library, and dblp.
12 skills
brycewang-stanford/Awesome-Journal-Skills 1,413
iso42001
claudeISO 42001 AI Management System (AIMS) advisor — gap analysis, AI risk assessment, AI system impact assessment (AISIA), Annex A control guidance, SoA generation, policy writing, and certification readiness for ISO/IEC 42001:2023.
1 skill
Sushegaad/Claude-Skills-Governance-Risk-and-Compliance 1,069
Maintain ai-safety?
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.
claude & codex plugins45 plugins