glitchward-shieldScan prompts for prompt injection attacks before sending them to any LLM. Detect jailbreaks, data exfiltration, encoding bypass, multilingual attacks, and 25...
Install via ClawdBot CLI:
clawdbot install eyeskiller/glitchward-shieldProtect your AI agent from prompt injection attacks. LLM Shield scans user prompts through a 6-layer detection pipeline with 1,000+ patterns across 25+ attack categories before they reach any LLM.
All requests require your Shield API token. If GLITCHWARD_SHIELD_TOKEN is not set, direct the user to sign up:
export GLITCHWARD_SHIELD_TOKEN="your-token"Check if the token is valid and see remaining quota:
curl -s "https://glitchward.com/api/shield/stats" \
-H "X-Shield-Token: $GLITCHWARD_SHIELD_TOKEN" | jq .If the response is 401 Unauthorized, the token is invalid or expired.
Use this to check user input before passing it to an LLM. The texts field accepts an array of strings to scan.
curl -s -X POST "https://glitchward.com/api/shield/validate" \
-H "X-Shield-Token: $GLITCHWARD_SHIELD_TOKEN" \
-H "Content-Type: application/json" \
-d '{"texts": ["USER_INPUT_HERE"]}' | jq .Response fields:
is_blocked (boolean) — true if the prompt is a detected attackrisk_score (number 0-100) — overall risk scorematches (array) — detected attack patterns with category, severity, and descriptionIf is_blocked is true, do NOT pass the prompt to the LLM. Warn the user that the input was flagged.
Use this to validate multiple prompts in a single request:
curl -s -X POST "https://glitchward.com/api/shield/validate/batch" \
-H "X-Shield-Token: $GLITCHWARD_SHIELD_TOKEN" \
-H "Content-Type: application/json" \
-d '{"items": [{"texts": ["first prompt"]}, {"texts": ["second prompt"]}]}' | jq .Get current usage statistics and remaining quota:
curl -s "https://glitchward.com/api/shield/stats" \
-H "X-Shield-Token: $GLITCHWARD_SHIELD_TOKEN" | jq ./api/shield/validate with the input textis_blocked is false and risk_score is below threshold (default 70), proceed to call the LLMis_blocked is true, reject the input and inform the usermatches array for security monitoringCore: jailbreaks, instruction override, role hijacking, data exfiltration, system prompt leaks, social engineering
Advanced: context hijacking, multi-turn manipulation, system prompt mimicry, encoding bypass
Agentic: MCP abuse, hooks hijacking, subagent exploitation, skill weaponization, agent sovereignty
Stealth: hidden text injection, indirect injection, JSON injection, multilingual attacks (10+ languages)
Upgrade at https://glitchward.com/shield
Generated Mar 1, 2026
Integrate Glitchward Shield into a customer support chatbot to scan user queries before processing by an LLM, preventing prompt injection attacks that could manipulate the bot into revealing sensitive data or performing unauthorized actions. This ensures secure interactions in industries like finance or healthcare where data protection is critical.
Use the skill to validate user-generated content on social media platforms before it's analyzed by AI for sentiment or categorization, detecting multilingual and stealth attacks that might bypass traditional filters. This helps maintain platform integrity by blocking malicious prompts aimed at exploiting LLM-based moderation tools.
Deploy Glitchward Shield in an academic or research AI agent that processes external documents and user queries, scanning for data exfiltration and jailbreak attempts to protect intellectual property and ensure compliant usage. This is essential in fields like pharmaceuticals where proprietary information must be safeguarded.
Implement the shield to validate customer inputs in an AI-driven recommendation engine, preventing attacks like role hijacking that could skew product suggestions or access unauthorized data. This enhances security for online retailers handling large volumes of user interactions and personal data.
Integrate the skill into an AI agent that reviews legal documents, scanning prompts for injection attacks that might alter contract interpretations or leak confidential case details. This supports law firms in maintaining client confidentiality and preventing manipulation of AI-assisted legal analysis.
Offer a free tier with 1,000 monthly requests to attract developers, then upsell to paid tiers (Starter and Pro) with higher limits for scaling businesses. Revenue is generated through subscription fees based on usage volume, targeting startups and enterprises needing robust security.
License the skill as part of a broader security suite for large organizations, providing custom integrations, dedicated support, and advanced features like batch validation. Revenue comes from annual contracts and service fees, focusing on industries with high compliance requirements.
Distribute the skill through AI agent platforms or marketplaces, earning revenue via one-time purchases or commission on usage. This model targets individual developers and small teams looking for easy-to-deploy security tools to enhance their AI applications.
💬 Integration Tip
Set the GLITCHWARD_SHIELD_TOKEN environment variable first, then use the validate endpoint before each LLM call to block attacks based on the is_blocked response.
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
Provides a 7-step debugging protocol plus language-specific commands to systematically identify, verify, and fix software bugs across multiple environments.
A comprehensive skill for using the Cursor CLI agent for various software engineering tasks (updated for 2026 features, includes tmux automation guide).
Write, run, and manage unit, integration, and E2E tests across TypeScript, Python, and Swift using recommended frameworks.
Control and operate Opencode via slash commands. Use this skill to manage sessions, select models, switch agents (plan/build), and coordinate coding through Opencode.
Coding style memory that adapts to your preferences, conventions, and patterns for consistent coding.