"Servers Offering MCP Prompts as Prompt Primitives" matching MCP servers:

Enables security researchers to evaluate AI system defenses against prompt injection attacks through a comprehensive set of test vectors and analysis tools.

Protects AI agents from prompt injection attacks, jailbreak attempts, and common web vulnerabilities by screening untrusted input through semantic LLM analysis and static pattern matching.

SkillTotal is a local stdio MCP server that lets an agent check a component before installing it. It statically analyzes MCP servers, agent skills, npm/PyPI packages and repos for supply-chain risks, dangerous capabilities (shell, filesystem, network, credentials), prompt-injection surfaces like tool poisoning, and exfiltration paths. Each finding cites file, line and snippet.

An MCP server that runs curated adversarial prompts against local Ollama models to test guardrails, scoring responses with heuristic verdicts for human review.

MCP server for auditing infrastructure-as-code attack paths, finding multi-hop chains from public internet to sensitive resources.

Enables AI coding assistants to scan projects for security issues such as leaked API keys, missing Supabase RLS, open Firebase rules, unauthenticated routes, and hallucinated packages, then fix and verify the results.

14 atomic MCP tools for AppSec and AI Security engineers: source/schema/prompt audit primitives, JWT inspect, HTTP diff, pentest atoms (default creds, GraphQL introspect, phpggc, interactsh OOB), and a defensive helpers library that fixes the bugs the detectors flag. SARIF output, PyPI Trusted Publishing with Sigstore provenance.

Provides security audit primitives for MCP servers, agent tool schemas, and system prompts, along with pentest utilities like JWT inspection and HTTP diffing.

Enables AI agents to scan MCP servers and AI packages for vulnerabilities, prompt injection, and supply chain attacks.

Enables crash-testing chatbots and AI agents from MCP clients by running multi-level audits and returning actionable reports with exact prompt improvements.

Gives AI agents threat-intelligence verdicts for IPs, domains, URLs and hashes, CVE lookups with EPSS and CISA KEV data, and a prompt-injection gate that scans untrusted text before the agent acts on it. Agents can also batch-check up to 100 indicators, hunt brand-lookalike domains, and track scan and request quotas.

Security scanning, prompt injection detection, secret leak detection, and agent permission auditing for AI agent workflows.

MCP server exposing TASIE's autonomous SAST, live exploit, and human-gated remediation as tools, enabling ChatGPT clients to scan code, dependencies, and frameworks and receive proof-carrying fixes.

Free deterministic security scan of a public git repository (GitHub, GitLab, Codeberg, Bitbucket): vulnerable dependencies via OSV.dev, secret patterns, and config lint, returned as structured JSON. Tools: scan_repository(url), audit_pricing(). Runs locally over stdio (python3 web/mcp_stdio.py or the Dockerfile) and is also hosted at https://project-feldspar.com/mcp. MIT licence, stdlib-only. Buil

Enables AI assistants to perform authorized web security assessments through MCP tools and a guided review prompt, including passive and full scans, WAF detection, remediation fixes, report export, triage, scan comparison, and proof-of-control verification.

Provides anonymous, privacy-preserving tools for AI agents, including prompt-injection scanning, site file validation, endpoint liveness checks, attestation, and risk-pattern analysis.

Scans MCP servers for prompt-injection, tool-poisoning, and SSRF vulnerabilities using 30+ canonical rules across 5 severity tiers, with optional signed safety reports for procurement.

An MCP server that exposes Prof. Null, an uncensored cybersecurity professor, as a set of specialized tools for learning hacking, CTFs, and security concepts through Q\&A, roadmaps, cheatsheets, labs, and more.

Enables plain-English website security checks from an MCP client, offering tools to assess site findings, explain individual issues, and list all available security checks.

MCP server that exposes 31 OSINT checks from Lissy93/web-check as tools for website analysis, including SSL, DNS, headers, WHOIS, and security presets. Enables natural-language-driven web recon and health checks.

Security scanning for AI coding tools (Claude Code, Cursor, Windsurf) including secrets detection, MCP config vulnerabilities, agent instruction checks, threat modeling, prompt injection testing, pre-commit security checks, and dependency vulnerability scanning.

An MCP server that provides safeguard capabilities to protect against prompt injection and unsafe tool calls.

Enables AI agents to perform professional penetration testing through a containerized Kali Linux environment, exposing industry-standard offensive security tools as structured MCP tools.

MCP server providing direct tool-access to security-testing primitives for bug bounty hunting, including recon, request replay, IDOR/BOLA fuzzing, vulnerability detection, secrets scanning, and persistent hunt memory with confidence-scored findings.

Enables deep security auditing of web applications directly from AI IDEs including Cursor and Claude Code. Scans URLs for vulnerabilities, returns security scores with SHIP/BLOCK verdicts, and provides specific fix prompts for remediation.

Exposes Tenable security operations as MCP tools for AI-powered security workflows, enabling asset, vulnerability, scan, plugin, and tag management via natural language.

A security scanner for AI coding agents and autonomous assistants that scans code for vulnerabilities, detects hallucinated packages, blocks prompt injection, and provides LLM-powered semantic code review via MCP or CLI.

A security research MCP server for testing tool call safety with deterministic policies like allowlists, path boundary enforcement, SSRF prevention, output redaction, and prompt injection detection, without requiring external LLMs or API keys.

Enables read-only security assessment of a public MCP endpoint by sending initialize and tools/list requests and returning a 0–100 score covering transport, auth, overly broad tools, injection phrasing, and rug-pull definition changes, plus a 12-probe prompt-injection suite against a bot you own that flags leaked system prompts, tool dumps, secrets, exfiltration channels, and canary obedience.

Provides local, dependency-free security scanning tools for LLM configurations, prompts, RAG sources, and more, enabling AI coding agents to detect prompt injections and other vulnerabilities without external network access.