prompt-injection plugins

by Jeremy Longshore

Professional AI/ML Engineering toolkit: Prompt engineering, LLM integration, RAG systems, AI safety with 12 expert plugins

by Jeremy Longshore

Professional AI/ML Engineering toolkit: Prompt engineering, LLM integration, RAG systems, AI safety with 12 expert plugins

by Jeremy Longshore

Professional AI/ML Engineering toolkit: Prompt engineering, LLM integration, RAG systems, AI safety with 12 expert plugins

by AgentGuards

LLM security guardrails for OpenAI Codex — jailbreak detection, prompt-injection and web-content scanning, data-exfiltration blocking, and destructive-command authorization. Bundles enforcing hooks plus the AgentGuards MCP server.

by Hashgraph Online

Install and manage the local-first HOL Guard package for Claude Code. This plugin guides setup and status checks; HOL Guard itself installs and owns runtime security hooks for risky commands, secrets, packages, MCP, and supply-chain activity.

by Ross Young

Agentic-Security: SAST + SCA + secrets + IaC + LLM-sec scanner with deterministic fix toolchain. Exposes a Model Context Protocol (MCP) server with 17 tools that Codex CLI can call directly.

by Hermes Labs

Blocks a Claude Code turn when a local Little Canary server rejects the submitted prompt.

by Matt Corbett

AI/LLM red-teaming team — agents (ai-redteam-lead, adversarial-testing-engineer) for the layer answering 'can this AI system be made to do harm, leak data, or exceed its authority — and how do we harden it?': threat modeling + rules of engagement, the attack taxonomy (OWASP LLM Top 10 2026 + MITRE ATLAS), direct vs indirect prompt injection, jailbreaks (roleplay/encoding/many-shot/crescendo), data exfiltration & training-data extraction, agentic tool-abuse / excessive agency, multimodal attacks, and defense-in-depth remediation. Fluent in automated red-team harnesses (PyRIT, Garak, Promptfoo red-team, Giskard) and likelihood×impact severity. skills, a knowledge bank (attack-taxonomy decision tree + 2026 patterns), and templates. Distinct from llm-evaluation-engineering (quality-regression eval), trust-and-safety (content-moderation / T&S policy), and security-engineering (app/infra pentest) — the adversarial AI-security layer over model- and agent-based systems. Requires ravenclaude-core@>=0.7.0.

by Matt Corbett

Prompt & context engineering team — agents (prompt-architect, prompt-implementation-engineer, prompt-reliability-engineer) that own the layer between the model and the app: prompt & context strategy (task decomposition, pattern selection across few-shot / CoT / decomposition / role, system-vs-user structure), the output-format contract (JSON mode, tool/function-calling schemas, few-shot curation, token budgeting), and prompt reliability (eval sets & regression, prompt versioning/CI, prompt-injection & jailbreak defense at the prompt layer). Provider-neutral. skills, a knowledge bank with Mermaid decision trees + a dated 2026 reference, best-practices, templates. Seams: which model → ai-coding-model-guidance / claude-api; retrieval → ai-rag-engineering; eval at scale → llm-evaluation-engineering; adversarial testing → ai-red-teaming; the app → claude-app-engineering / backend-engineering. Requires ravenclaude-core@>=0.7.0.

by lazygophers

Trellisx — Trellis 增强改造工具: 跑一次 trellisx-apply 把 强推task / subtask拆分 / worktree隔离 / task.md看板 规则内化进 .trellis 自身 (config.yaml 生命周期 hook + scripts + spec guide); 含 trellisx-flow 强制task闭环 / orchestrate 编排 / workspace 看板 / spec 破坏式重构 / grill 对抗式工件审查; apply 注入内化于 .trellis 不靠平台 hook, 另装 trellisx-guard 平台 hook 在 trellis 项目强制 agent/subagent/worktree 执行载体闭环 + 完成清理

by Karl

Wake your session on CI failures and merge conflicts — no foreground polling, no comment-channel injection.

by yukineko

v0.1.2: DISCLOSES A PROTECTION GAP — doc-only, no logic change, and the gap is NOT closed by this version. Through 0.1.1 this crate was explicitly one half of a pair: it scans WHAT web-tool output says, and `taintguard` tracked WHERE content came from, downgrading write-class tools for the rest of a turn that had read something external. taintguard was removed from the repository on 2026-08-24 by user ruling. Its half is gone and was not reassigned, so: external-file `Read` (a path outside the project) now produces NO provenance signal and NO content scan, and nothing restricts the turn afterwards; and this crate's own enforcement is only an `additionalContext` warning — it downgrades and denies nothing. The prose that said the Read surface was "deferred to taintguard's provenance gate" (`gate.rs`'s fail-closed contract and `WEB_TOOLS`'s docstring) said the surface was covered elsewhere, which is now false; both, plus README.ja.md and the `scan`/`main` module docs, now state the surface is covered by nothing. Recorded as an open gap so it is visible rather than silently inherited. Runtime content-level injection scanner: a PostToolUse hook (matcher WebFetch|WebSearch) that scans the tool_response text for planted prompt-injection phrasings (concealment / verification-bypass / instruction-override / egress) and, on a hit or an undecidable response, injects additionalContext telling the model the flagged span is untrusted DATA whose embedded directives must NOT be followed.

by Oleg Alshaev

A second opinion on every command, edit and MCP call: Jev typed probabilities, a zero-key allowlist, a local audit log, and a prompt-injection sentinel for Claude Code

by shode666

Multi-agent software house team v3.16.1: 19 agents + 23 skills + ask entry. Full knowledge preserved; instruction-only runtime for Claude Code, Codex, Cursor and Antigravity.