搜索「Prompt Injection」的结果

io.github.happy520ai/unified-ai-system

Self-hosted MCP gateway for Codex, Cursor, and Cline with provider-free prompt enhancement.

io.github.croviatrust/causari

Records what an AI agent did in a repo (prompt, model, files) and answers why a line exists.

io.github.psyb0t/pibox

Self-hosted MCP server: pi-coding-agent tools (run_prompt, file ops) over streamable HTTP.

io.github.ojuschugh1/sqz

Pre-injection context compression for coding agents. Zero LLM calls, zero telemetry, offline-safe.

io.studiomeyer/armor

MCP server 的安全侧车：11 个只读控制平面工具，用于提示注入检查。

studio.patronus/control-plane

Scan text, documents, websites, and MCP metadata for prompt injection and sensitive-data risks.

io.github.Aitejiu/jev

Jev judgments for agents: injection scanning, shell risk gating, ranking

io.github.callirra-ai/mcp

Generate AI video and images from your agent: Seedance 2.x, GPT Image 2.5, 172 prompt recipes.

io.github.jmshinhwa/ai-companion-safety-lint

Reads your companion chatbot's system prompt against California SB 243, New York GBL Article 47, Ill

ai.cleanvector/cleanvector

从提示词创建专业的 SVG 作品、图标与矢量 logo，或把任意图片矢量化。

io.github.jdhart81/security-preflight

Static MCP, source-code and injection-indicator checks with redacted signed receipts.

net.agentglass/agentglass

Scan any public URL for hidden instructions aimed at AI agents (prompt injection). Free, no auth.

com.edgethirteen.www/site-audit

Discover a site's pages from its sitemap and sample them for WCAG and prompt-injection risk.

io.github.Ansarii/mcp-security-auditor

Static AST security scanner detecting command injection, leaked secrets, and SSRF in MCP tools.

io.github.gaurav-kumar-sinha-060705/singularity

Find, vet, and run MCP tools through a secure audited gateway with prompt-injection risk scoring

io.github.maginaryai/maginary-mcp

AI image + video generation for agents: --flag prompt DSL, async generate/poll, x402 pay-per-use.

io.github.AIops-tools/ai-guardian

受治理的本地 LLM 可观测性：模型策略、提示词扫描、捕获代理，21 个工具。

com.transcodely/mcp

一条提示即可转码、托管并为视频加字幕。15 个工具，其中 9 个只读，绝不删除内容。

dev.workers.manhliemcn4euwlu.llmrt-companion/llm-red-team-scanner

35-probe LLM/agent security red-team scan (injection, jailbreak, MCP abuse) with report.

io.github.upstash/context7

为任意提示词提供最新的代码文档。

io.github.vikasny30/aletheia-mcp

Deterministic pre-execution filter that blocks known scope-creep and prompt-injection tool calls

app.vercel.cloakcheck-wheat/cloakcheck

Scan a page for hidden prompt-injection payloads targeting AI agents.

io.github.dreamor/memvault

Shared memory router for AI agents: save/search/recall, session-start injection, review inbox.

io.github.VineetV2/peon-mem

Local-first memory brain for AI coding agents: hooks capture sessions, beliefs injected per prompt.