Parry — Data skill for Claude Code
by Dmytro Onypko - Prompt injection scanner for Claude Code hooks.
How to install Parry
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open vaporif/parry-guard and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Parry does
[](https://github.com/vaporif/parry/actions/workflows/check.yml) [](https://github.com/hesreallyhim/awesome-claude-code)
Alternatives in Data
- Context Mode — Benchmark Results — Benchmarked against real outputs from popular Claude Code MCP servers, Skills, and dev tools 5.6k ★
- LLM Redteam — LLM red-team corpus runner — fires categorized prompt-injection / jailbreak / system-prompt-leak / data-exfil 4.5k ★
- 06 Ink React Terminal UI — Prompt 06: Verify and Fix the Ink/React Terminal UI Pipeline 2.3k ★
README
Parry-guard
[](https://github.com/vaporif/parry/actions/workflows/check.yml) [](https://github.com/hesreallyhim/awesome-claude-code)
Prompt injection scanner for Claude Code hooks. Scans tool inputs and outputs for injection attacks, secrets, and data exfiltration attempts.
**Early development** — this tool is under active development and may have bugs or false positives. Tested on linux/macOS.
Prerequisites
The ML models are gated on HuggingFace. Before installing:
- Create an account at huggingface.co
- Accept the DeBERTa v3 license (required for all modes)
- For
fullmode: also accept the Llama Prompt Guard 2 license (Meta approval required) - Create an access token at huggingface.co/settings/tokens
Usage
Add to `~/.claude/settings.json`:
**With [uvx](https://docs.astral.sh/uv/):**
{
"hooks": {
"PreToolUse": [{ "command": "uvx parry-guard hook", "timeout": 1000 }],
"PostToolUse": [{ "command": "uvx parry-guard hook", "timeout": 5000 }],
"UserPromptSubmit": [{ "command": "uvx parry-guard hook", "timeout": 2000 }]
}
}
**With [rvx](https://github.com/vaporif/rvx):**
{
"hooks": {
"PreToolUse": [{ "command": "rvx parry-guard hook", "timeout": 1000 }],
"PostToolUse": [{ "command": "rvx parry-guard hook", "timeout": 5000 }],
"UserPromptSubmit": [{ "command": "rvx parry-guard hook", "timeout": 2000 }]
}
}
**With parry-guard on PATH** (via [Nix](#nix-home-manager), cargo install, or [release binary](https://github.com/vaporif/parry/releases)):
{
"hooks": {
"PreToolUse": [{ "command": "parry-guard hook", "timeout": 1000 }],
"PostToolUse": [{ "command": "parry-guard hook", "timeout": 5000 }],
"UserPromptSubmit": [{ "command": "parry-guard hook", "timeout": 2000 }]
}
...
Related Skills
Pipeline Manifest
Map research scripts to inputs, outputs, dependencies, figures, and tables,
Yotta Agent Hardening
YuanSafe (元安全) — defensive hardening workflow for AI agents: static config-facing scan of an agent's own runti
AgriQuant AI
AI-powered weather intelligence for agricultural commodity futures. Predicts price moves 48-72 hours early acr
Exfiltration Triage
Judge whether outbound content or an egress connection is a real data-exfiltration attempt.
Note To Artifact
Use when turning notes, data, or a plan into a beautiful, self-contained HTML artifact (report, dashboard, pla
Scanner Feature
Develop a new scanner feature using the architect-to-CI pipeline
Related Agents
Lens Security
Security lens of the production readiness audit. Attacks the codebase on paper - injection, SSRF, path travers
Tp Audit Scanner
Use this when the user asks for a security / quality audit, pre-release sweep, or "scan this for issues". Find
Secrets Auditor
Credential scanner — scans staged diff and working tree for secrets, tokens, and credentials. Groups findings