Promptrejectormcp — AI skill for Claude Code
Prompt Rejector protects your AI-powered applications from prompt injection attacks, jailbreak attempts, and traditional web vulnerabilities.
How to install Promptrejectormcp
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open revsmoke/promptrejectormcp and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Promptrejectormcp does
Prompt Rejector protects your AI-powered applications from prompt injection attacks, jailbreak attempts, and traditional web vulnerabilities.
Alternatives in AI
- Hol Guard — Open-source antivirus for AI agents: block risky tools, secret access, prompt injection, malicious packages, M 485 ★
- Prismor — Self-hosted runtime control plane for AI agents 334 ★
- Engineering Notebook — A Bun CLI that ingests Claude Code and Codex session transcripts, generates LLM-powered daily summaries, and s 320 ★
README
Prompt Rejector
**Screen prompts, skill files and MCP tool descriptions before an AI agent acts on them.**
Prompt Rejector combines deterministic security checks, focused **TypeSafe Jev judgments**, and a configurable reasoning model. The active configuration uses TypeSafe and Gemini; Claude and OpenAI adapters are available through model configuration.
| Use it from | Connection | Start command |
|---|---|---|
| An application or script | HTTPS API: https://localhost:3001 |
npm start |
| Codex / Claude Code | Plugin + MCP over stdio | Install the plugin or configure the Node launcher |
| Claude Desktop chat | Bundled .mcpb extension + skill |
Desktop installation |
| ChatGPT / Claude web | Skill + tunnel or authenticated remote MCP | Web connection guide |
The local API and MCP connections can run together and use the same analysis configuration. Prompt checks use **`POST /v2/check-prompt`**. `/v2` is the only current API; `/v1` is retired. MCP keeps all 11 tool names and needs no version selector.
**Already using Bryan's local installation?** The HTTPS service and Codex MCP entry are configured. Go to the [local service runbook](https://github.com/revsmoke/promptrejectormcp/blob/codex/typesafe-model-routing/docs/operations/local-server.md#installed-macos-service) for status and restart instructions; do not start a second server on port 3001.
Related Skills
Nab
Token-lean web microfetch for LLM agents: any URL → clean markdown via CLI, MCP server, and Claude Code plugin
Autopentest AI
Agentic Pentesting MCP server that discovers, exploits, and reports web application vulnerabilities.
Humanizer Hk
去 AI 味 · Strip the AI tells out of written Chinese. Hong Kong Traditional. Claude Code skill + paste-anywhere
Bagman Skill
Secure key management patterns for AI agents - storage, session keys, leak prevention, prompt injection defens
Agent Skills Deep Dive
A Python implementation of the Claude Agent Skills system — a sophisticated prompt-based meta-tool architectur
Prompt Architecture
Reference guide for designing production LLM prompt systems in this codebase. Covers the Two-Pass Gap Analysis
Related Agents
Kavach LLM
KAVACH LLM/AI security specialist. Audits prompt injection (direct + indirect/RAG), jailbreak/persona/system-p
Review Security
Reviews code changes for security vulnerabilities including injection attacks, sensitive data exposure, insecu
Security Architecture
Boucle implements a defense-in-depth security model to protect against prompt injection attacks and maintain t