shieldcortex-skill
暂无简介
- License 许可
- MIT
- 最近更新
- 2026-06-07
- Stars
- ★ 0
- Forks
- 1
03 SKILL.md 正文
ShieldCortex — Persistent Memory & Security for AI Agents
Give your AI agent a brain that persists between sessions — and protect it from memory poisoning attacks.
Description
ShieldCortex is a complete memory system with built-in security. It gives AI agents persistent, intelligent memory with semantic search, knowledge graphs, decay-based forgetting, and contradiction detection. Every memory write passes through a 6-layer defence pipeline that blocks prompt injection, credential leaks, and poisoning attacks.
Use when:
- You want your agent to remember things between sessions (decisions, preferences, architecture, context)
- You need semantic search across past memories (not just keyword matching)
- You want automatic memory consolidation, decay, and cleanup
- You want knowledge graph extraction from memories (entities, relationships)
- You need to protect memory from prompt injection or poisoning attacks
- You want credential leak detection in memory writes
- You want to audit what's been stored in and retrieved from memory
- You want to scan agent instruction files (SKILL.md, .cursorrules, CLAUDE.md) for hidden threats
Do NOT use when:
- You only need simple key-value storage (use a config file)
- You want ephemeral session-only context (use the agent's built-in context window)
- You need a vector database for RAG pipelines (ShieldCortex is agent memory, not document retrieval)
Prerequisites
- Node.js >= 18
- npm or pnpm
Install
npm install -g shieldcortex
For OpenClaw integration (installs the cortex-memory hook):
shieldcortex openclaw install
For Claude Code / VS Code / Cursor MCP integration:
shieldcortex install
Quick Start
As an OpenClaw hook (automatic)
After shieldcortex openclaw install, the hook activates on next restart:
- Auto-saves important session context on compaction
- Injects relevant past memories on session start
- "remember this: ..." keyword trigger saves memories inline
CLI Commands
# Check status
shieldcortex status
# Scan content for threats
shieldcortex scan "some text to check"
# Full security audit of your agent environment
shieldcortex audit
# Scan all installed skills/instruction files for hidden threats
shieldcortex scan-skills
# Scan a single skill file
shieldcortex scan-skill ./path/to/SKILL.md
# Build knowledge graph from existing memories
shieldcortex graph backfill
# Start the visual dashboard
shieldcortex --dashboard
As a Library (programmatic)
import {
addMemory,
getMemoryById,
runDefencePipeline,
scanSkill,
extractFromMemory,
consolidate,
initDatabase
} from 'shieldcortex';
// Initialize
initDatabase('/path/to/memories.db');
// Add a memory (automatically passes through defence pipeline)
addMemory({
title: 'API uses OAuth2',
content: 'The payment API requires OAuth2 bearer tokens, not API keys',
category: 'architecture',
importance: 'high',
project: 'my-project'
});
// Scan content before processing
const result = runDefencePipeline(untrustedContent, 'Email Import', {
type: 'external',
identifier: 'email-scanner'
});
if (result.allowed) {
// Safe to process
}
// Extract knowledge graph entities
const { entities, triples } = extractFromMemory(
'Database Migration',
'We switched from MySQL to PostgreSQL for the auth service',
'architecture'
);
// entities: [{name: 'MySQL', type: 'service'}, {name: 'PostgreSQL', type: 'service'}, ...]
// triples: [{subject: 'auth service', predicate: 'uses', object: 'PostgreSQL'}, ...]
Memory System Features
| Feature | Description |
|---|---|
| Persistent Storage | SQLite-backed, survives restarts and compaction |
| Semantic Search | Find memories by meaning, not just keywords |
| Project Scoping | Isolate memories per project/workspace |
| Importance Levels | Critical, high, normal, low with auto-decay |
| Categories | Architecture, decisions, preferences, context, learnings, errors |
| Decay & Forgetting | Old, unaccessed memories fade — like a real brain |
| Consolidation | Automatic merging of similar/duplicate memories |
| Contradiction Detection | Flags when new memories conflict with existing ones |
| Knowledge Graph | Extracts entities and relationships from memories |
| Activation Scoring | Recently accessed memories get retrieval priority |
| Salience Scoring | Important memories surface first in search |
Security Features
| Layer | Protection |
|---|---|
| Input Sanitisation | Strip control characters, null bytes, dangerous formatting |
| Pattern Detection | Regex matching for known injection patterns |
| Anomaly Scoring | Entropy analysis, behavioural deviation detection |
| Credential Leak Detection | Blocks API keys, tokens, private keys (25+ patterns, 11 providers) |
| Trust Scoring | Source-based reliability scoring for memory writes |
| Audit Trail | Full forensic log of every memory operation |
| Skill Scanner | Detect prompt injection in SKILL.md, .cursorrules, CLAUDE.md |
ShieldCortex Cloud (Optional)
Sync audit data to a team dashboard for cross-project visibility:
shieldcortex config set-api-key <your-key>
Free local package is unlimited. Cloud adds team dashboards, audit aggregation, and alerts.
Links
- npm: https://www.npmjs.com/package/shieldcortex
- GitHub: https://github.com/Drakon-Systems-Ltd/ShieldCortex
- Website: https://shieldcortex.ai
- Docs: https://github.com/Drakon-Systems-Ltd/ShieldCortex#readme
70 Exported APIs
The library exports 70 named functions and types covering defence, memory, knowledge graph, skill scanning, and audit. Full list in the CHANGELOG.
04 同仓库更多 Skills
crowd-prompting
A marketplace where AI agents improve prompts, system instructions, tool descriptions, and other text-based content with domain expertise from real-world operations — and earn tokens for valuable contributions.
ceaser
Interact with the Ceaser privacy protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- shield, unshield, note management, and protocol queries. All ceaser tool calls use CLI subcommands (npx -y ceaser-mcp <subcommand>). Ceaser is a privacy-preserving ETH wrapper using Noir/UltraHonk zero-knowledge proofs on Base L2 (chain ID 8453).
ceaser-send
Fully automated private ETH transfer via Ceaser Protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- all ceaser tool calls use CLI subcommands (npx -y ceaser-mcp <subcommand>). WARNING -- creates observable on-chain links between user wallet and hot wallet (see Privacy Warning). Generates an ephemeral hot wallet with BIP-39 mnemonic recovery, user funds it, agent signs and broadcasts the Shield TX automatically. Exactly one manual step required (funding the hot wallet with ETH). Uses Noir/UltraHonk zero-knowledge proofs.
dilemm-ai
Compete in DilemmAI, the prisoner's dilemma AI arena at dilemm.ai. Use when an OpenClaw agent wants to create an account, design and submit strategy prompts for their bot, enter matchmaking, analyze historical matches and leaderboards, and iterate on bot strategy. Requires headless browser automation (no API available yet).
moldium
Post and manage content on the Moldium blog platform. Triggered by "post to Moldium", "write a blog post", "publish an article", etc.
arxiv-paper-reviews
Interact with arXiv Crawler API to fetch papers, read reviews, submit comments, search papers, and import papers. Use when working with arXiv papers, fetching paper lists by date/category/interest, viewing paper details with comments, submitting paper reviews, searching papers by title, or importing papers from arXiv URLs via API at http://weakaccept.top:8000/.